We were quite pleased with the results of our first week of BP Idol. The panel generally found the submissions worthy of praise.
“I love the construct here. He really draws me in from the first sentence and then takes me through the process in a clear fashion.” — Will Carroll on Ken Funck
“The study was interesting, and I liked the fact that you took it to a second level after explaining how small the differences where.” — Kevin Goldstein on Brian Oakchunas
“Matt did this really very well, with a confidence in his prose that speaks well as far as what we might anticipate his subsequent work.” — Christina Kahrl on Matt Swartz
Of course, into every contest week a little rain must fall, and not all reviews trended so positively.
“unlike his initial, winning entry, the light humor from before is missing” — Kahrl, on Matthew Knight
“She’s clearly not as comfortable with the numbers and it shows here.” — Carroll, on Brittany Ghiroli
“I’m not sure this one worked for me as well.” — Goldstein, on Jeff Euston
In the end, voting was fairly close, but Byron Lescroart narrowly becomes the first elimination in the Finals of the contest, despite having a thumbs-up percentage better than two other entries. Byron, you got a lot of good comments and comfortably graded out in the five percent of the initial entries. Well done, and thank you for coming this far with us.
Your remaining contestants
Questions or comments about the contest thus far? Leave them in the comments section, rate the best ones up via the plus/minus buttons, and we’ll review the favorites for inclusion in future BP Idol content.
Just wanted to say thank you to everyone who read and posted on my two articles, and of course to the judges for selecting my initial entry. This was a great learning experience for me, and my first real attempt at writing for any sort of publication online or otherwise, so certainly there are lessons to be learned and improvements to be made. I am, of course, disappointed to have been voted out - especially with this coming week's topic being fantasy baseball - but, alas, "the people have spoken."
Would I have done things a little differently with the Basics article in retrospect? Yes and No. Yes, I probably would have looked to address the tone of the article to make it more engaging, but at the same time I love the concept of DIPS theory and the history behind it, so at least will not regret my choice of topic. Most importantly, however, I think I made the mistake of underestimating the time and effort required to publish the caliber of article you are used to seeing here at BP, and that's the kind of mistake that will get you kicked off the island.
Each of the remaining writers is wonderfully talented in his or her own way, and I wish them all nothing but the best of luck going forward in this competition and beyond. I will be eagerly awaiting the next batch of articles with the rest of you, and look forward to seeing many of you in the comments threads.
So, thanks again to everyone who participated, the judges, and especially BP as a whole for coming up with what I think is, if nothing else, a thoroughly entertaining concept. And with that I bid adieu...
Byron, you did a great job and I definitely want to see more of you in the future. Out of hundreds of submissions, you were one of the ten, and no one can ever take that away from you. In the meantime, remember to tag "BP Idol Finalist" on your resume. It might open up some doors for your career and/or your writing pursuits.
Meanwhile, after a bit of a breath, make sure to stick around and follow the rest of the competition. You might pick up some pointers. I know I have :)
Appreciate it, Richard, I really do. Also, I am excited to continue to follow the competition immediately and will look forward to your comments on the response threads for all future articles.
Slightly related, I have to mention the "power rankings" of sorts that you made for the articles. For lack of a better word, I thought those were awesome. Just awesome. When I saw them I laughed and was telling my roommate about the fact that you had made them, and he laughed back at me and said "Byron, I think you are the only other person I know who would 100% do something like that." The statement is, of course, completely true. I generally write full power rankings for the fantasy leagues I'm in with my friends, so to see someone make a run at doing that here was, in my opinion, phenomenal. Just about everyone I know loves to read/discuss power rankings, even if they don't actually like to put them together. Here's hoping we see another set next week!
I'm glad you liked them :) I had the idea about four hours before voting closed, but wanted to wait as late as possible since I didn't want it to significantly influence the voting. It's also nice to know you and your roommate read and liked them. I was expecting something like that to get -feedback, actually. But it was fun to do and I plan on keeping it up unless it upsets BP or the other readers.
I mean, if I was an authority on all this stuff, I would've been a finalist right? :)
Thanks, Brian. I've certainly enjoyed your writing as well.
I love PitchF/X, and I'd definitely encourage you and all of the other remaining finalists to consider incorporating it into an article if you are comfortable with it. In my opinion, PitchF/X is an incredible tool that can really be used to "re-simplify" the game of baseball. A significant part of what baseball is all about is the showdown between the pitcher and the batter, and PitchF/X, amongst other things, can provide amazingly detailed data on exactly what went down in that showdown. It allows us to use pictures/graphs to analyze hitters and pitchers at a very high level without requiring a PhD in statistics. Anyone can interpret PitchF/X data with a little nudge in the right direction, and most people who haven't got a clue what advanced statistics are can garner incredibly advanced insight into the "science of baseball" by utilizing PitchF/X as a tool to bounce theories off of.
Example: "Oh, this guy has a terrible O-Swing% leading to a below average LD% and FB% costing him power, which, given his body shape projections doesn't bode well." Huh? But now we can just pop up a little chart of the strike zone with different symbols representing different pitches and explain "see those dots? He's swinging at breaking balls that end up there, in a tough place to hit, and not swinging at fastballs that seem to go right down the middle. This is a problem as there's no way we can hit a curveball in the dirt very hard, and fat guys who hit soft grounders everywhere aren't ideal..." At the same time, we're getting even more information here than just O-Swing% data because we are also learning what TYPE of pitch the hitter is having problems with. It's a silly example, but I think it speaks to how I could literally get my 12-year old, baseball fanatic cousin to relate to what's going on in PitchF/X, and thereby really help grow his understanding of the game without forcing him into reading a bunch of decimal points and percentage signs. This is also true when looking at the PitchF/X dealing with release points of pitchers as well - if the dots are all over the place it's not good... KISS theory reborn?
This is just one small example of something we can do with PitchF/X. I really think we're looking at a "PitchF/X revolution" in the analysis of baseball as we know it, and it's best to jump on that wave while there's room at the front.
Absolutely agree. Here's what I'd do to make it work:
We already have the technology to track movement and velocity from a certain point, right? So first I'd secure this data for balls as they come off the bat. If we've got direction, movement, and velocity, we should be able to plot a suspected course for the batted ball in question. Then, if we could get an aerial shot of the field, complete with GPS coordinated fielder locations displayed on the background of the field itself, we should be able to calculate the time each fielder has to get to each batted ball at a certain point. If we then take a look at what percentage of balls the fielder is able to get to and successfully convert into outs we can run that data against the "HitF/X" data off the bat to determine how "difficult" it was to really make that play based on the time the fielder has to get to the place where he made the play, and the distance he had to cover to get there.
This is a rough concept, but I think it's possible - with current technology - and I'd bet some smarter teams already are putting this type of analysis into effect.
Stats Inc has a system that divides up the field into grids and analyzes hits that way... though I last saw the system about 10 years ago so there might've been some refinement since then.
I don't know if GPS will give the amount of precision needed though, but I do like the idea of using the movement and velocity of the ball as it comes off the bat though since bats are made of wood and have some slight imperfections, the angle might be imprecise.
It means that the percentage of people that read his article and then voted for him was higher than a couple others. That said, it's the raw number of "thumbs up" votes that decides who stays and who goes.
Hello again! I just wanted to chime in here because I woke up today to see myriad responses up on this post clamoring about the injustice of the elimination process. Since I was, in fact, eliminated, I figured I'd toss my two cents into the fountain. My take is simply this: we knew the rules of the game before we played, so there's no sense complaining about them after we've started. While I suppose the "justice" of the eliminations is an interesting theoretical debate, it really should be no more than that. The guys (and gal!) running the show here were clear about the rules before the voting started so I want to be clear now that in no way, shape, or form am I "upset" by the manner in which I was eliminated this week. I'm not sure why my article didn't receive as many views as others did, but I'm quite sure it wasn't as a result of a diabolical BP scheme to force me out.
I prefer to look at it this way: if I had written a better article, I'd still be writing.
Again, I greatly appreciate all of the support from the subscribers, it's been an absolute thrill for me just to be involved to this point.
Finally, to play off a phrase I think I remember reading in one of Dave Studeman's articles that I linked to, rate stats are great, but counting stats aren't completely worthless either: "there's value in dem dar numbers..."
I'm not sure that your logic is really true. "I prefer to look at it this way: if I had written a better article, I'd still be writing. "
Being on the west coast, I didn't have nearly enough time yesterday to get to the all of the articles. I used first the title and then the pictures to determine which articles I'd read. I picked yours because at that moment I wanted to read something about FIP, but your picture was unfortunate. It looked like it was cropped from a picture taken outside of a frat house. I'm going to venture a guess and say that someone was doing a kegstand within 50 feet of where that picture was taken?
Point is, I think that might've led to less people choosing your article as one of the few they read.
If they'll only give us a couple of hours to read these articles and make good judgments, I think the selection criteria is f'ed.
It sounds like plenty of people read each article. In fact, Byron only needed 10 more people to vote for his article. Seeing that only 19% of those who read his article voted for it, there were more than enough people that may have voted for him had he "written a better article".
#1 the 10th place vote getter who got 19% of reads
#2 another vote getter who had at least 20% more reads than each other finalist and a "significantly higher" vote total than 10th place but had a voting percentage lower than 19% of reads
#3 the 1st place vote getter who got 41% reads
I can't really guess things the number of the second lowest reads, because it could be just one more read than the lowest vote getter.
Nor can I guess how many reads the second greatest reader got (to calculate case #2).
I also can't guess how much a "Lower percentage" is, but I'll assume rounding might be at play in Will's description to arrive at 19% and 41%, so I will assume that case #2 is at 18% (so that it is less than case #1)
Since I can't tell specific number of reads, merely guess, I will assume #1 and #3 got the same amount of reads and #2 got 20% more reads than #1/#3.
I will also assume that the number of thumbs up has to be large enough where a difference of less than ten thumbs up between 9th place and 10th place seem "close". The number of thumbs up also has to be large enough so that the difference between case #2 and case #1 is "significant"
If there are 5000 reads:
Case #1 has 950 thumbs up votes (5000*0.19)
Case #2 has 1080 thumbs up votes and 6000 reads (6000*0.18)
Case #3 has 2050 thumbs up votes (5000*0.41)
With a difference of 130 votes between case #1 and case #2, it is possible for 10th place at 950 votes and 9th place at a number less than 960 votes to appear "close".
Seems like a lot of reads and votes... but if you try the same scenario with just 1000 reads:
Case #1 has 190 votes and Case #2 has 216 votes, which is not a "significant" gap.
I would guess that fewer people actually read his article, and so he had the fewest tallied thumbs-up votes. However, the percentage of people who read his article and gave it a thumbs up was higher than other finalists.
I guess titles are going to increase in importance now... or maybe BP readers were turned away by a non-BP statistic? I surely hope that wasn't the case...
I think he means that they wouldn't recognize the stat as not being a BP stat. If the stat works, and provides insight, I don't really care where it came from. BP didn't invent HR either, but we still refer to them frequently.
1. Byron had the lowest total of votes ("counting stat") of the ten finalists. This was (and will be) the only criteria for eliminating a contestant. You get more votes, you stay. The lowest vote total goes. There's no input from judges or anyone else at PEV (aside from their own votes, which count as much as yours.)
2. Byron was less than ten votes below the #9 total.
3. No one got 50% of votes.
4. Byron got 'thumbs up' from 19% of people that read his article.
5. One contestant with a significantly higher vote total had a lower percentage, largely because that entry was read by 20% more than the next most-read entry. (No, I don't know why either.)
6. The highest vote total got thumbs up from 41% of readers, but oddly had the second lowest 'read count.'
7. The vote totals were stratified, with a clear "top three", a clear "bottom three", and then the middle four grouped tightly.
Finally, I'd like to thank Byron for his participation. Getting this far is quite the accomplishment and I think it speaks well to his potential.
I have a REAL problem with the method of choosing the eliminated writer. As has been pointed out by many, reading this number of articles in a such a compressed time period is a lot to ask. Given that, there is going to be variation in the number of reads which may have nothing to do with the writers themselves.
This is still BP right? The site where performance is almost always measured in terms of a rate vs number of opportunities. Seems like an odd choice to abandon that philosophy for your own writing contest. Shame on you...
I admit to opening all of the entries in browser tabs and not getting the chance to read them all. BP marked them all as read, but only about 7 of 10 actually were. It's a silly way to tally votes. Why no thumbs down option?
Enticing headlines are important. Look at how popular Twitter is becoming. I don't click on every BP.com link, nor do I click on every ESPN.com/mlb link or CNNSI.com baseball link. I click on something I think will interest me, whether it's an interview with a player I like or a concept I'm interested in learning about... I can be biased with the best of them.
However, enticing headlines, photos, flashy bios or graphs will only get a person so far.
And it's not voting him off... it's voting for other people more... which is a big difference.
I get the percentage thing. Obviously significantly fewer people read Byron's article than two other entries that had a smaller number of more votes. Some people may have chosen not to read it due to its subject or due to their not liking his initial entry enough to bother reading this one. That's fair.
What I don't get is "graded out in the five percent of the initial entries". Does that mean 10 < 5% of total entries i.e. the total number of entries was just over 200?
I didn't get that comment either... I thought it was some roundabout way of saying that his initial submission scored better than the other finalists with all three judges.
Sorry for the lack of clarity there. I was saying that to make the top 10 Byron had to be comfortably in the top 5% of initial entrants. We had hundreds of entries.
The same is true of our other nine continuing contestants. We're very proud of this group of people, mean to say.
Well, um, from prior information from Will, we knew there were hundreds of applicants... so if there were at least 200 applicants, then the top 10 would have to be in the top 5% by default. 200 X .05 = 10
Still a nice compliment... but there was probably a better way to phrase it overall.
Wait. Am I missing something? You posted their articles on Memorial Day (5/25) and closed the voting the next day? I thought I had more time than that.
PLEASE announce on the article in several places the day the voting closes. I read through the article and this is what it said:
"Each week, you're be the voting body that decides who went big and who goes home. The process is simple. During the voting period, you'll have the opportunity to vote for any entry you feel is worthy—just click the thumbs-up icon at the top right of each article display."
Evidently, however, the information was in the black bar just above the ad... that I always ignore.
It would have killed you to put the info in the text of the article itself? It would have killed you to give more time to the voters to read 10 articles? This isn't American Idol, where all I'd have to do is listen to them yodel for 3 minutes. This takes longer than that.
This is one subscriber who is unimpressed with the producers of BP Idol right now.
Actually, the articles were posted Sunday mid-day-ish and voting closed Tuesday night, so it wasn't "the next day".
There was a BP Unfiltered post with a direct link, a main article link post on the BP.com page and a black bar at the top... then you also get the BP Subscriber email... The BP Unfiltered post noted that voting was closing soon, the bpidol link mentioned what time voting closed and the black bar had a countdown... are more links to the article and more indications of the countdown needed?
I think more time would have been nice, especially for those who only have time to read BP articles in general on the weekend... but then that would extend the length of the contest or force a finalist each week to write an article only to find out they were eliminated. Maybe if the articles were posted Friday, voting closed Monday night and the deadline for submissions was Wednesday...
The first time I saw a deadline was in the "you have several hours left" post from Will while I was at lunch at work. The black bar at the top, yes, I might have caught that, but inclusion in the articles themselves is something that I'd expect. I'm sorry that I routinely ignore the "subscriber emails" because mostly those are teasers for what's on the site, which I already visit.
If you're referring to this article: "May 25, 2009 - Prospectus Idol: Round One - The Basics by Dave Pease" then you should note the date and the fact that the deadline is not listed in that article. If you're referring to "Meet the Finalists" from 5/17, there's no mention of a deadline. If you're referring to the April 12th article, BP Idol, there's no deadline listed.
The broader point is that I was promised "there's also going to be a lot of opportunities for Baseball Prospectus subscribers in general to make a significant difference. By using your power as the voting public to evaluate and vote for the competitors themselves, you will help shape the future content on the site." But how's that going to happen in half a Sunday, Memorial Day, and Tuesday? That's functionally one day or, if you have no family commitments on Memorial Day weekend, 2.5 days.
That's not nearly enough time to "participate actively by commenting on the articles themselves, making a case for which ones you like and why, and helping sustain a debate on the questions these articles will inevitably raise." Why bother when that debate will be so truncated?
What I was referring regarding the deadline to was the section of the http://www.baseballprospectus.com/bpidol/ page where, just beneath the title "Idol 2009 Round One: The Basics", it now says "Voting for Idol 2009 Round One: The Basics has closed." Before it changed, it had the deadline for voting.
I agree it would be nice to have additional time, particularly to check the reliability and accuracy of some of the math and the cited sources.
They were actually posted Sunday midafternoon, but ... we know it was a lot of articles to read in a short period of time. We want this to run weekly, though, and the winners have got to have some time to do their next entry, so that limits our options.
We'll get the voting period in all the Idol materials from now on, though. Sorry about that.
I too am a little concerned with the timeframe at the beginning of the contest, also. Ten articles is a lot to digest for people with a job, a wife and three kids. I only got to three articles because of work commitments. I hadn't gotten to Byron's entry yet. Is there some rush to get this done in ten weeks?
I knew the time frame, but I agree that it was way too short. I read four of the articles very closely and carefully. That took a few hours, especially because I also commented on them. I read the other six but with not nearly enough time to give them the scrutiny (or comments) they deserved. That's part of the reason that I voted for 7 of the 10--I didn't think it would be fair to reject an article to which I wasn't able to pay proper attention.
I'll agree on the timeframe. I know you guys don't want to drag this thing out forever, but I can't see giving less than 5 days if you're talking about 10 articles. Obviously the number of articles will decrease each week, but this first week was overwhelming. I'll be completely honest - I got through about two articles and I chose not to vote for any because I didn't think it would be a fair thing for me to do.
I know it's silly for me to be complaining about a contest (an idea I love, by the way), and that my voting/participation is voluntary, but I just want to give some feedback for future decisions.
Isn't equally unfair to give some authors a free pass to the next round because you didn't read them, and reject others who you did read? I think if you vote without having adequately inspected each article, then there is some unfairness. But in the short time frame, that might still be the best option.
"Byron Lescroart narrowly becomes the first elimination in the Finals of the contest, despite having a thumbs-up percentage better than two other entries"
So in other words, you're putting more weight on counting stats than on rate stats? Clearly, you didn't read Ken Funck's article...
Or did the judges assume the roles of the 2000 US Supreme Court???
I am also surprised that the percentage doesn't weigh heavier than the vote total. Or that I didn't know this before hand. Luckily I did read all 10 articles (which showed up pretty early in the day on Sunday actually). Still, I would not be too happy if I were Byron.
I don't think you can have the percentage weigh heavier than the vote total... or else you run into cases where an article that has 20 thumbs up out of 50 reads sticks around when an article with 199 thumbs up out of 500 reads gets dropped.
I wanted Byron to stick around more than some of the other finalists... that being said, proofreading would've helped him grab a few extra votes.
Isn't the question, why is this article being read 500 more times? Does that have to do with its randomly generated placement (probably not)? The picture of the person writing it? The title of the article?
I would rather have the higher percentage article because of the people who actually took the time out of their day to read said article, they actually enjoyed it.
Is BPro Idol trying to generate voting for the best article or the writer who brings the most subscriptions to the website?
It could have to do with the picture of the person writing it, or the title of the article. Hey, it's the internet... most people browsing the web, who might not be taking the contest as seriously, need to be grabbed in the title or they'll move on to something else. I can see why BP would appreciate that.. they're running this contest to see who can garner enough attention to click on an article link.
Secondly, if it's the raw thumbs up count, and thumbs up votes can only be given by subscribers, then people who are not subscribed but like a particular BP Idol finalist might just subscribe just to vote for their favorite person... and that is also another kind of author that BP wants.
I mean, sure, BP wants to advance knowledge and enjoyment of baseball and all that stuff.. but they want to make money too.
Remember, this competition is about which author people are going to pay to read. Sometimes that might not be the "best" article, but it might be the most "entertaining" article. As for myself, I find things that excite my creativity or make me think about baseball in new or different ways entertaining. Others prefer jokes, still others prefer regression analysis. On the flipside, I discontinued my ESPN.com membership.
Ideally, I would think (and I may be wrong) that the winner of this contest would have some measure of many strengths... a good, entertaining writer with a grasp of sabremetrics and original, creative ideas to research.
We have to keep in mind that there was an introductory phase as well. Those articles probably had a significant impact on who was read this time around.
For example, I know my very first article read this round was Ken Funck, simply because I enjoyed his initial piece. I then went (roughly) in order of who I wanted to read based on my enjoyment of their introductory pieces. I did get around to all the entries eventually, but I could just as easily not got around to all ten.
Much respect to Byron on his work here. It's quite good. Still, there seems to be an awful lot of hand-wringing over a contestant who couldn't get 1 out of every five readers of his article to give him the thumbs up.
I'm really surprised by both the elimination method and the voting window. You guys took how long to decide on the finalists and then you expect your readership to make an elimination decision in two days? Wow...
But then I've been really unnerved by how this has turned out so far, especially with the selection of established writers. I really have to agree with Tango's take on the competition.
This is how I feel. You guys had weeks (months?) to narrow your selections down. Granted, you had many, many more than ten to go through. But I would think that would at least clued you guys in to how much time should be spent reviewing each submission.
At the very least it's really refreshing to see how seriously the subscribers here are taking this competition. We all pay for this website and love it dearly, and want to make sure the best new writer is selected.
I disagree; It makes it easier for me to essentially 'double vote' for one contestant by voting for one and then simply clicking through the rest and not voting.
Having only sub'ers vote makes it easier to detect potential gaming, but doesn't eliminate the possibility.
Yes, (i) the time allotted for reviewing the submissions is woefully inadequate, (ii) the fairness of the scoring methodology is arguable, and (iii) life ain't fair. But having this contest in play sure beats the alternative (i.e., not having interesting articles from new writers to read). I'm giving the BP Idol crew a thumbs up for this undertaking. And I'm looking forward to the last few weeks of the contest when I'll have the time to give each writer's submission the attention it deserves. Until then, I'm relying on a more basic/juvenile/superficial evaluation approach: was I entertained or not?
I have no problem with the way the votes are being tallied, but it sure was a short window to read them. Bergstrom is saying these articles were posted "mid-day-ish" on Sunday... well he must be on the west coast. The articles were posted after 5pm EDT. I was checking all day, had nothing to read while I was watching the game, then found them posted after I had no more time. What I don't understand is this: If the articles were to be submitted by noon on Friday, and they receive no more edits from that point, what takes so long to post them? If it takes over 48 hours for the judges to read and comment on them (AND this is their JOB and they must allow time for this activity in their schedules,) why do we get less time than that? I managed to read only six of the ten. I knew the deadline; it WAS clearly marked on the top of the Idol page, but I did not have the time to give equal and fair consideration to all articles. Maybe the holiday got in the way of giving proper time in this particular week, but I do not have a desk job so I'm not sure if it would make that much of a difference for me personally. I would respectfully request that the Idol entries be posted at or around noon EDT on Sundays, just like the weekday articles.
How about posting the articles on Saturday - when we faithful readers do not other BP articles to read? Given the high stakes of this contest, it would behoove the administration to consider revising some of their methods to help ensure that we voters have a greater chance of reading all of the submitted entries. I read them all, and it left me with very little time for other activities in the voting window. As I knew 9 writers would advance, I decided to grade both the initial entries and the first submission and combine the scores, voting for all but the bottom score. Byron was not the bottom score, and so got my vote. That said, I did not feel so strongly about his essays to be seriously upset by the elimination - I applaud his writing and his achievements in making it in when so many others (including yours, did not). My bottom line is, if I get a chance to choose among a pool of analysts/writers for a regular column, I want to choose the person who has the most to offer me. I would hope for an opportunity to be able to make a clear, rational decision when choosing. Good luck Byron, in your future endevours, writing and other. And to the remaining nine contestants, good luck in the next round.
I think a lot of people would have complained about something no matter who was voted out of the competition. The volume of complaints seems low enough to not need to worry about it.
Who knows what it was - unintersting title, didn't like the look of the picture, found the subject matter boring. There were 10 articles to choose from in a very narrow time window - I doubt most of us could accurately describe why we picked the articles we did read - unless we simply went in order.
The fact is though that those articles with a lower percentage of yes votes managed to draw far more eyes to their page - that means more ad revenue potential for BP and that is an important if unsatisfying critiera to being a successful columnist.
It doesn't matter how well you write if people never bother to read it. For whatever reason - and it may not even have been something he could control - he couldn't draw sufficient eyes to stay in the competition.
I get your point, but it does seem that the complaints are consistent... those who favor a percentage approval among the articles which were actually READ are looking for some fair compensation for the fact that we were not given sufficient time to consider all articles. The voting period will obviously be less of a factor as the contest moves along, but I'm not sure that "The volume of complaints seems low enough to not need to worry about it."
Was I imagining it, or did the order of the articles listed on the main Idol page change as I was reading them... perhaps to reflect the vote tally? This might be helpful, but does give more chances to those who had already received more votes, assuming most people just went right down through the list like I did. If they do change, any chance you can adjust the order factor to "least read" instead to give a more equal viewing to each article? Articles with the best titles might still get the most reads, as they should.
Is there any particular reason why the process could not use a two-week window rather than the current one-week? I agree with the main concern which is that we weren't really given enough time to digest the articles. Operating with a 2 week window would allow 1 week for readers and a week for the survivors to prepare their next article.
I am only complaining about the timeframe to vote. It was a holiday weekend and we were away without internet access until late Monday. Just a little extra time to review them at this point would be nice. As the field gets winnowed down, it will be easier to read them all, but this week was not particularly fair to Byron (or whoever would have been eliminated).
That said, I think Christina, Kevin and Will deserve a ton of credit for getting this going!
Good effort, Byron. I liked your article better than many of the others, but what do I know? I thought Kris Allen would be one of the first ones voted off as well.
Here's the good news, folks - You only have 9 articles to read next week. I get the feeling that we'll have a worthy winner in the end.
Overall, with the unexpected high volume of submissions with only three people (i.e. the judges) reading through them coupled with introducing a new engine that has columnist pictures, tallies reads and votes and then the forethought to add in a randomizer, I think this has system has been very well implemented so far. I give extra points because this is the first time BP has done this and all things considered, there has been relatively few growing pains. The site hasn't crashed from increased traffic, articles didn't mysteriously disappear and votes didn't get magically cleared. Thus, kudos. I imagine it'll be even better the next time around as they'll have a better idea what to expect. Being able to read each article will also be easier as the number of finalists decrease.
Heck, I can't even post a comment to the ESPN boards without relogging in as a member (thus losing my comment) and again waiting for the java-lag to get done. I can't even access CNNSI's boards from work without a bunch of formatting issues. The Rockies had millions of dollars behind them and couldn't even sell World Series tickets correctly. So, give 'em a bit of credit.
As American Idol now has four judges, I think Baseball Prospectus would be best served with four panelists. Three from BP, plus one regular reader to give an outside perspective on how a viewer would receive these prospective columnists. I nominate Richard Bergstrom for the position.
Very true. And others have been doing the same, as I think they should be. The way I look at it is if there's only going to be one left, then the least I can do is try to help out the other nine... and hopefully do that without someone hitting me with a trout.
Wow. I gave Byron's piece a thumbs up. There were two other submissions that I thought were of lesser quality than his, though a couple I liked better including Hissey's also on FIP.
Good luck to you, Byron, in all your future endeavors.
Byron's piece was easily my favourite of the first round. As I said in the comments, I wish I could have voted for it twice.
I'm very disappointed to see him voted out. I was very much hoping he would be the ultimate winner. BP has, I think, become quite insular in its use of stats (your continued use of Clay's old defensive numbers in recent years, when they'd clearly been surpassed), and Byron specifically citing perhaps the best analysts working today - who simply don't happen to work for BP - was a breath of fresh air.
Well, it is a popularity contest, not the Pulitzer... that being said, it's not like there will be a bunch of dummy accounts abusing the voting system since only paid subscribers can vote.
I actually thought Jeff Euston's picture had the worst resolution, etc. Doesn't really matter to me if they're there or not.. I would prefer that their bio be reposted with every article so I don't have to duck back to the initial entry to find their background.
I don't think he was referring to the resolution quality. The picture was a poor choice for the competition.
I think bio's should be hidden. The articles should stand on their own. I wouldn't mind if they were anonymous. If you believe what they wrote when they announced the contest then you wouldn't want this to turn into a popularity contest.
Taking the American Idol concept a little too far IMO.
Right on the second point. If we don't have enough time to read all of the submissions then people will use whatever they can to prioritize. The picture was an awful choice for the contest. The piece's title was also a little lame.
For the first point, I'm not convinced that wouldn't be better solved by changing the method by which people are kicked. Probably too late to change that though.
Yes. I am a Business Intelligence Reporting Analyst for a college, have a Masters in Education and am pursuing a PhD... but I also multitask and it takes some time for my VBA code to run and generate reports. I also sing karaoke, go to concerts, read books, and other non-computer stuff. Oh, and yeah, I got to about 15 baseball games a year.
Oh, and I have long blonde hair and a beard, which means my mugshot looks like I'm a stoner or an ex-convict... though when I was thinner (and less bald on top) I got more Jesus comparisons.
Wow, didn't realize how many people complain about this, as if they have something at stake here. This is not un fair to anybody since everyone in the contest follows the same rules. You may have only been able to read 5 articles, but so did someone else and they are at random.
I think the articles thet get read the most are the ones with the most comments and most likely the most controversial. If I had time to read 1 article, and I saw one had 57 comments and another had 4, I'd read the one with 57 comments. If you are boring, then people don't read your stuff. It's a fact of life and it works fine.
Interesting thought. Maybe some people are just opening articles to look at the comments and not voting for or against. As a contestant, I am not certain that BP's choice of using the counting stat is the wrong way to go. If someone opens my article but doesn't really read it and then doesn't vote, I wouldn't want it to count against me.
And currently, it doesn't count against you... at least, assuming the number of thumbs-up votes are different. It might matter more (or less) in a tie breaker.
No idea if there are tiebreakers or what they are... but the thumbs up percentage or the number of readers could be used as a tie breaker if it came down to it.
Yeah, I read them all, and voted, but then re-opened a few of them later to read the new comments. It's people like me that are why you can't use votes/view! (Incidentally, does refreshing the page count as a new view?) Seems like the problem could have been avoided with both Thumbs Up and Thumbs Down buttons, but honestly I don't think it will be much of a "problem" really -- people with a low percentage likely aren't going to last long.
You write "You may have only been able to read 5 articles, but so did someone else and they are at random." Then immediately contradict that by pointing out that there's an information cascade. I don't believe the 5 articles people click on are random. I read 6 and I didn't choose randomly.
We do have something at stake - the winner becomes a regular columnist - I want to ensure that BP will get a new columnist that I will want to read on a regular basis.
There is almost no way that the contest delivers the best new columnist. It's going to deliver someone well-rounded who can write about a number of different subjects but it isn't going to produce the person who would be the best new columnist.
Many of the current BP writers couldn't win a contest like this. Their areas of expertise and their writing styles aren't as well rounded as these contestants need to be.
I would think a "dedicated specialist" who wrote well would be able to land himself a pretty good job without this contest. There are a lot of ways to get your work out there.
Nate, Will and Kevin all started by dabbling in specific areas and developed a fan base.
I disagree that the winner of this contest is going to be a well-rounded columnist without a voice or specialty. I wouldn't say that all nine finalists who advanced handled the first topic well. Of course, I wouldn't say that those who voted did so for the articles that met the topic objective, BP Basics, well, either.
Anyone who writes does so in his/her own voice. I would assume that the ones who advance are the ones who write about each week's topic in a clear, easy, and entertaining way using his/her voice.
I'll agree and disagree. First, I probably wouldn't have won this contest. I can't do the hardcore stats and I'm not a great writer. With that said, I think the person that wins will be well-rounded, but will have their own voice and stand out. This *is* a popularity contest, but not an experiment in any way. We realized that we didn't know and couldn't find all the writers that had talent, in much the same way that record companies didn't find Kelly Clarkson or Chris Daughtry. Maybe it won't be your cup of tea, but it will be a winner who will have earned it.
But as for not winning, someone pointed out how I got my job here. Or how Nate did. Or Kevin. Or more recently, Shawn Hoffman. This is just an alternate path using a fun method. People are really, really overthinking this.
I've got no problem with what you are doing, I was just saying that there is a chance you won't end up with the best writer because the best writer might get knocked out early by something like fantasy baseball.
I will say that if you want people to take the time to read these articles and give any thought to who they are voting for then their critique of the scoring system should matter to you.
What's interesting is that Kelly Clarkson and Chris Daughtry weren't *Kelly Clarkson* and *Chris Daughtry* before Idol, and wouldn't be who they are without Idol. I wonder if BP Idol will work the same way.
They're calculating this thing right. RAw totals is the correct way. Let's me clear - it's not the % read because no one knows whether people actually read it. It's the % merely opened.
So calculating this based on who has the highest % votes among opened is a completely meaningless percentage.
It's not clear cut, you're right, but without the numbers you can't really say whether that problem is more onerous than people not being able to read all of the articles. I would suspect the latter is a bigger issue.
In any case you're too far inside the box on this one. The method can be changed to avoid a tradeoff like that. Adding the option to give an article an up or down vote would get rid of the "i opened a tab and didn't read the article" issue.
Not really, because there will be articles people will read and feel neutral about, so won't care to click up or down. And besides, I think it'd be better if people suggest ways for others to improve instead of justifying why they're thumb-downing/flaming an article.
Oh, and if people are already having problems reading all the articles, an additional vote click option can slow things down for some as they ponder whether it's a thumbs down or a neutral.
Hi all. In response to several points you've raised:
* Elimination: let's not over-engineer the process here. If for whatever reason you like an article and/or contestant and want them to continue, click the thumbs-up. You can read the articles in random order from the landing page, or you can pick and choose articles based on whatever criteria you choose.
* Timeframe: I'm sorry we didn't do a better job of communicating the voting deadline for this first week, and we'll do this on a comprehensive basis in the future.
As far as length of voting period, we know you might not have time to read everything the first two or three weeks, but we're not planning on changing it. The winners need to know they're working on an article with a hard deadline.
The contest needs to be weekly. Both BP and our partners are looking for the contestants to produce a content on topics that might not be of their choosing on a regular basis by the playoffs this year. There's a huge difference between someone who can produce a few excellent articles on topics of their choosing a year and someone who can do it on weekly assignment. We love to work with both types but we're looking for the latter in this contest.
We didn't think of it early enough this year to give much more time even if we wanted to. Next year, if we run a contest like this again, we'll be more organized.
* Headshots: everyone's always telling us "hey, if you all are professional sportswriters, you ought to have headshots." We could put in a switch that can change them to smilies like we have on the subscription page if that'll help.
116 comments have been left for this post.
Just wanted to say thank you to everyone who read and posted on my two articles, and of course to the judges for selecting my initial entry. This was a great learning experience for me, and my first real attempt at writing for any sort of publication online or otherwise, so certainly there are lessons to be learned and improvements to be made. I am, of course, disappointed to have been voted out - especially with this coming week's topic being fantasy baseball - but, alas, "the people have spoken."
Would I have done things a little differently with the Basics article in retrospect? Yes and No. Yes, I probably would have looked to address the tone of the article to make it more engaging, but at the same time I love the concept of DIPS theory and the history behind it, so at least will not regret my choice of topic. Most importantly, however, I think I made the mistake of underestimating the time and effort required to publish the caliber of article you are used to seeing here at BP, and that's the kind of mistake that will get you kicked off the island.
Each of the remaining writers is wonderfully talented in his or her own way, and I wish them all nothing but the best of luck going forward in this competition and beyond. I will be eagerly awaiting the next batch of articles with the rest of you, and look forward to seeing many of you in the comments threads.
So, thanks again to everyone who participated, the judges, and especially BP as a whole for coming up with what I think is, if nothing else, a thoroughly entertaining concept. And with that I bid adieu...
Byron, you did a great job and I definitely want to see more of you in the future. Out of hundreds of submissions, you were one of the ten, and no one can ever take that away from you. In the meantime, remember to tag "BP Idol Finalist" on your resume. It might open up some doors for your career and/or your writing pursuits.
Meanwhile, after a bit of a breath, make sure to stick around and follow the rest of the competition. You might pick up some pointers. I know I have :)
Appreciate it, Richard, I really do. Also, I am excited to continue to follow the competition immediately and will look forward to your comments on the response threads for all future articles.
Slightly related, I have to mention the "power rankings" of sorts that you made for the articles. For lack of a better word, I thought those were awesome. Just awesome. When I saw them I laughed and was telling my roommate about the fact that you had made them, and he laughed back at me and said "Byron, I think you are the only other person I know who would 100% do something like that." The statement is, of course, completely true. I generally write full power rankings for the fantasy leagues I'm in with my friends, so to see someone make a run at doing that here was, in my opinion, phenomenal. Just about everyone I know loves to read/discuss power rankings, even if they don't actually like to put them together. Here's hoping we see another set next week!
I'm glad you liked them :) I had the idea about four hours before voting closed, but wanted to wait as late as possible since I didn't want it to significantly influence the voting. It's also nice to know you and your roommate read and liked them. I was expecting something like that to get -feedback, actually. But it was fun to do and I plan on keeping it up unless it upsets BP or the other readers.
I mean, if I was an authority on all this stuff, I would've been a finalist right? :)
I'm curious in hearing more about the time and effort you took.
Byron, I enjoyed your stuff and thought your work with Pitch F/X was great. Good luck in the future!
Thanks, Brian. I've certainly enjoyed your writing as well.
I love PitchF/X, and I'd definitely encourage you and all of the other remaining finalists to consider incorporating it into an article if you are comfortable with it. In my opinion, PitchF/X is an incredible tool that can really be used to "re-simplify" the game of baseball. A significant part of what baseball is all about is the showdown between the pitcher and the batter, and PitchF/X, amongst other things, can provide amazingly detailed data on exactly what went down in that showdown. It allows us to use pictures/graphs to analyze hitters and pitchers at a very high level without requiring a PhD in statistics. Anyone can interpret PitchF/X data with a little nudge in the right direction, and most people who haven't got a clue what advanced statistics are can garner incredibly advanced insight into the "science of baseball" by utilizing PitchF/X as a tool to bounce theories off of.
Example: "Oh, this guy has a terrible O-Swing% leading to a below average LD% and FB% costing him power, which, given his body shape projections doesn't bode well." Huh? But now we can just pop up a little chart of the strike zone with different symbols representing different pitches and explain "see those dots? He's swinging at breaking balls that end up there, in a tough place to hit, and not swinging at fastballs that seem to go right down the middle. This is a problem as there's no way we can hit a curveball in the dirt very hard, and fat guys who hit soft grounders everywhere aren't ideal..." At the same time, we're getting even more information here than just O-Swing% data because we are also learning what TYPE of pitch the hitter is having problems with. It's a silly example, but I think it speaks to how I could literally get my 12-year old, baseball fanatic cousin to relate to what's going on in PitchF/X, and thereby really help grow his understanding of the game without forcing him into reading a bunch of decimal points and percentage signs. This is also true when looking at the PitchF/X dealing with release points of pitchers as well - if the dots are all over the place it's not good... KISS theory reborn?
This is just one small example of something we can do with PitchF/X. I really think we're looking at a "PitchF/X revolution" in the analysis of baseball as we know it, and it's best to jump on that wave while there's room at the front.
Now we just need HitF/X and we can produce perfect defensive metrics.
It's coming...
Absolutely agree. Here's what I'd do to make it work:
We already have the technology to track movement and velocity from a certain point, right? So first I'd secure this data for balls as they come off the bat. If we've got direction, movement, and velocity, we should be able to plot a suspected course for the batted ball in question. Then, if we could get an aerial shot of the field, complete with GPS coordinated fielder locations displayed on the background of the field itself, we should be able to calculate the time each fielder has to get to each batted ball at a certain point. If we then take a look at what percentage of balls the fielder is able to get to and successfully convert into outs we can run that data against the "HitF/X" data off the bat to determine how "difficult" it was to really make that play based on the time the fielder has to get to the place where he made the play, and the distance he had to cover to get there.
This is a rough concept, but I think it's possible - with current technology - and I'd bet some smarter teams already are putting this type of analysis into effect.
Stats Inc has a system that divides up the field into grids and analyzes hits that way... though I last saw the system about 10 years ago so there might've been some refinement since then.
I don't know if GPS will give the amount of precision needed though, but I do like the idea of using the movement and velocity of the ball as it comes off the bat though since bats are made of wood and have some slight imperfections, the angle might be imprecise.
What I meant to insinuate but didn't really succeed at was that it is very much in the works and will be here sooner rather than later.