[ed – 2017: while slightly outdated, much of this article still applies. The images are new, replacing those which were in the original article]
Every spring since launching the Player Forecast Manager, we at Baseball Prospectus have tried to improve it to make it more useful for our subscribers. The PFM, for those who haven't tried it, is a tool that uses our Depth Chart projections to create player values for use in fantasy baseball drafts. It's popular among our users—enough that its use has occasionally bogged the entire site down in the past—but its potential has still not been fully realized.
Since last year, the PFM has been given a major overhaul by Ben Murphy, Rob McQuown, and others, who entirely rewrote and streamlined the code to improve the speed of the program. The PFM itself was also moved to a separate server so that it would run faster (and keep it from slowing down the rest of the site), and it was given a new coat of paint that is part aesthetics, part user interface. Today, we'll run through some PFM basics, while explaining the additions we've made, and some still to come.
The Player Forecast Manager—which you can reach through the new, easy-to-remember address of pfm.baseballprospectus.com—gets its strength from customization. Is your league rotisserie, or points style? Are you using a standard 12 teams, or just six? How about 30? Maybe you play in an AL- or NL-only league, rather than mixed. All are customizable options. Even if your league uses a complex scoring system that involves statistics like RBI + R + HR, or Saves + Holds, or Quality Starts, you can set the PFM to base its calculations on those as well.
Once your league settings are in place, the PFM will spit out dollar values and draft rankings that reflect your choices. For instance, there are often discussions about how someone like Adam Dunn would be worth more in a league that uses on-base percentage—the PFM can tell you exactly how much the difference is. (For reference, the answer is $21 for standard, $28 for OBP leagues.) If your league uses two starting catchers and five outfielders, the values for those positions will change, as more of them are required—Logan Morrison is worth $5 in an OBP league (with everything else standard), but in a league with six outfielders and 14 teams, he's worth $10, thanks to new scarcity and replacement levels.
Those are the base values, but they can change as a draft progresses. This is why we include inflation. Let's say the first five outfielders in this hypothetical OBP league are all auctioned off at $40 apiece as the first five players, which is somewhere between $8 and $12 more than they are worth. The PFM will recognize that there is less money in the pool and that there are fewer outfielders to choose from, and will adjust prices accordingly. In this scenario, Hanley Ramirez's price drops by roughly 10 cents, but as a draft goes on (and depending on how high or low you set the inflation influence), those numbers will continue to drop, assuming people continue to overpay. Conversely, if players all go for less money than expected, more money will be in the pool to spend on the remaining players and their values will rise accordingly.
If you want more details about these different areas of customization, check out our detailed help page, which you can also find in one of the orange buttons atop the new PFM. And that's as good a segue as any for getting into the meat of the shiny new things the PFM can do for 2011.
For one, those orange buttons, which are very much not hidden:
The help page, as mentioned, is new. The "News & Updates" button also leads to a new page, which details the changes within the PFM and which version we are currently running. (For example, the last major change was made to version 2.2 of the PFM on February 23, when (H+BB+HBP) was added as a scoring category.) The two buttons at right export spreadsheets with all of the data contained within the PFM, minus the calculations for customization. Think of it as a depth charts projection spreadsheet. (You can also export whatever customized setup you output to the PFM, for further sorting/customization/offline usage.)
Beneath those buttons is another new area for saving your inputs. Maybe you're tired of adding in the same stat customizations for your league every single time you enter the PFM, or you don't want to continue to bookmark each league and its various rules in order to visit them later. You don't have to now, as you can save your setups and return to them at any time—they are saved to your account, and so as long as you are signed in to Baseball Prospectus, you can access the data. I, for one, am thrilled that I don't have to revamp the offensive categories every time I want to check prices for players in one of my personal leagues, and we hope that you get the same kick out of being lazy.
You may have noticed the Configuration Options section looks very different as well:
See those little blue circles with the "i" in the middle? Those are tooltips. Mouse over the circle, and a message will appear detailing what that specific option does. Observe:

You don't have to keep checking the detailed help section if you don't want to, and the mystery surrounding some of these options has been removed.
["Show Expert Ratings" is no longer available, but these images should help explain how to mark a player as "taken"]
Second base is an example of why this is a useful service—while PECOTA and the PFM see Brandon Phillips as an impact second baseman, I'm a bit more pessimistic, and would prefer any of the three second basemen who follow him in the PFM rankings. This doesn't mean you have to agree with me, but it does show you that this player's 2011 isn't an open-and-shut case. By clicking on the "Tier 3" text in Phillips' row, you can open up the second base rankings where I commented on him, and see why I'm more pessimistic than the PFM is. (We hope in the future to make these comments available within the PFM simply by mousing over them, just like the tooltips.)
Another new addition can be found under the "League Used" setting under "League." For those of you who play in custom fantasy leagues that don't employ any of the mixed, AL- or NL-only options—maybe you only pick from certain teams, or divisions, or what have you—we now have you covered, as you can choose any combination of 30 teams, and the PFM will offer output only on players from those selected clubs. (Testing for this option made me want to play in a six-team league that uses just American League East clubs.)
Those are the major changes—you can read about the smaller ones in the News & Updates section—but it's not the end of them. We have one final addition, which is the player health ratings produced for the Team Injury Projection series. In the past, these were released as their own spreadsheet, but now, just like the Expert Rankings, they can be switched on in the PFM and exported. The PFM will now allow you to see its own customized valuations, my fantasy rankings, and the health ratings of every player covered, all in one place on your draft day.
The Team Injury Projection series (formerly the Team Health Reports) are an important fantasy baseball cornerstone, and they will be coming in their new format soon, courtesy of Corey Dawkins and myself. The health ratings for those players will appear in the PFM before they release in article form, for those who need that information for their drafts. That's courtesy of our brand new injury projection system, which you will be hearing much more about in the coming weeks and months.
There, in a 1,600-word nutshell, is your 2011 PFM. (Just as a reminder, the PFM, depth charts, and injury risk ratings are available to both fantasy and premium subscribers.) We won't rest there, though, as we have more ideas for future versions of the program, and are always taking feedback from you, the reader. Let us know in the comments what you think about the new PFM, and what else you would like to see implemented in the future. We're always listening, and always tinkering.
Thank you for reading
This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.
Subscribe now
My draft is over, but please get us upside scores ASAP. That, in my dynasty league, is immensely helpful for trade evaluation. Would have been helpful for the draft, but, you know, spilled milk and crying and all that.
PFM is useless to anyone who doesn't have internet access at their draft.
In fact, I was invited to join a new league last year that specifically prohibited internet access during the draft and considered my request to bring my laptop very suspiciously. (I was dumbfounded as well.) It's a very old league with several 'older' members who aren't technologically savvy. The draft lasts 10-12 hours long and they were concerned that I would have access to fresh injury/trade/cut news during the draft.
Even if internet access were allowed, last year's draft was held in a cinder-block community rec room. No internet available unless you had 3G.
I'm one of them. Our league drafts at a clubhouse in one of the owners subdivision.
We draft there because Its very large with plenty of table space and room. The drawback, it has no internet access.
For example, in my 10 team league, I set the minimum value at $0, and ran the PFM. It outputs 10 catchers, but it would be nice to see who it thinks #11 would be. I tried putting the minimum value < $0, but this seemed to make the whole thing give weird results.
I know I can look in "depth charts" or in the fantasy rankings by position to find out who #11 might be, but it would be easier if more players were automatically included in the spreadsheet.
If you change the minimum league salary, that will impact valuation and not the output.
However, I have 10 batters and 5 pitchers (with 10 batters openings {1B,2B,3B,SS,OF,OF,OF,C,Util,Util}) and I still have batters in the output. I understand that the batters may still have more 'value' than the pitchers left, but if I say that I only have 10 batter spots, PFM should recognize that and stop showing them.
Currently the PFM does not anticipate your roster positions enough to be able to discern whether players you have taken fill your roster or not. So even though you may have filled your roster for hitters, it will still evaluate hitters based on the overall league budget and money available.
Clearly it would be useful for PFM to be able to anticipate your roster (and your league mates' rosters as well). Ideally it will be something that is customizable where you can ask the PFM to consider either stats, roster positions, or both when using user-centric inflation. As a result of this complexity, these features are on our road map for inclusion into the valuation updates in 2012.
One thing I do miss, though, is the feature that told you what your max bid was based on your roster size and players remaining. It's not that hard to figure out on my own, but it was a nice touch.
The improvements more than balance this out for me, though.
A trade evaluation tool sounds terrific.
Also, I would love to be able to import the inflation feature to my computer -- I was forced to abandon it when I tried using it in the past because my auction league is far too fast-paced. With the code modifications and migration to a stand-alone server, I'll probably give it another shot.
I don't see any reason to project:
Tommy Hanson 143.3 proj. IP
Josh Johnson 140
Jonathan Sanchez 138
While at the same time projecting:
Johan Santana (!!!) 210 IP
Javier Vazquez 201 IP
Livan Hernandez 199.7 IP
Daric Barton 619 PA vs. Jose Bautista 548 PA vs. Josh Hamilton 515 PA
It would be helpful to be able to ignore PECOTA's projected playing time factors and display the data based on per PA/per AB producion. Yeah, I can do that in an Excel sheet but that's less handy.
love the changes.
one idea..is there away to take a player out of the pool w/o giving him a salary?
my league allows guys to put reserves on active roster after draft, so it doesn't count during auction.
great job!
And isn't, cannot be, one bit more effective than its underlying data base. In this case, the Depth Charts. Which, for one thing, project that 14 out of 15 last-place Seattle Mariner pitchers will finish with records over .500 That 15th one, the negatively-VORPed Lucas French, it projects for a .500 record.
I'll grant you progress, in that previous year's Depth Charts were often horrible. The current one is OK. But it appears you ignore them unless/until something DRAMATIC! happens, a la with Wainwright. You've still got a few dozen clear errors in them, a la projecting 160 games out of Pablo Sandoval.
Fixing the Depth Charts would accomplish something of substance.
Has anything changed in PFM that would account for this? What is that? Thanks.
Where are the player card updates (incl. 10-year forecasts, percentile forecasts and upside projections)?
Why are the projected records of teams different between Depth Charts and Post-season Odds pages? (Are they using different schedule assumptions; if so, how/why?). No, Colin's article (http://www.baseballprospectus.com/article.php?articleid=13040) does not answer these questions.
Why are some of the projected team records from PS Odds Report different from the "Exptected Win Pct" on the same report? E.g., Cards expected win pct. is .522 and expected wins is 85.8; Brewers expected win pct. is higher (.524) but expected wins is lower (85.2).
Why are the projected win-loss records of individual pitchers so far off from what they should be? Wins are being massively inflated, as has been pointed out by several users.
How often and how aggressively are the Depth Charts being updated? (Jason Castro, e.g., is still being projected with 400 PA, ~48 hours after a season-ending injury).
All of these questions are not new; rather, they have been raised repeatedly over the past several weeks by a wide range of users. I have yet to hear a definitive answer from anyone at BP -- either via customer service email, or via comments.
What about the PFM cards? In my keeper league, I decide how long to give a contract to a player based, in part, on how the player cards tell me he is likely to age. It seems that there's been tremendous investments in other parts of the BP universe, but the player cards are not yet ready, and many, many keeper league keepers are past due.
I don't know how much you use market research and focus groups to understand your customers' needs, and I, obviously, have no access to that proprietary information. But, if you do use these comments as a source of understanding what the customers want, please consider this a resounding request to focus on the needs of fantasy players. There are lots of great contyent streams on BP, but I suspect that many pay the $35/year because of PECOTA, PFM, and player cards (i.e. the fantasy stuff). Please, get that right early.
The PFM seems to be better than ever. And, in reading through the comments, it looks like there are bigger changes to come, which is great.
I do have a request, would it be possible to add 2B3B (double and triple total bases) as a category? Currently, I produce two spreadsheets, one with total bases as a category and the other with doubles and triples as separate categories.
From there I try to figure out which is the better projections.
Thanks, Brian
Would really help for picking breakout/rebound candidates.
Meanwhile, no more players were revealed. (The lower the adjusted min-salary the higher the top salary settings.)
Great product. How are the PFM values going to be updated throughout the year? Will they be projections for season totals as they change, or for remaining season results?
Thanks again!
Thanks,
Matt
When someone else takes a player, check the left-most box in the inflation column.
When YOU take a player, check both boxes in the inflation column.
Then, when you hit update, those players will be added to the "your team" and "other teams" taken lists, as appropriate. It's actually pretty cool.