Submit chat questions for Jeffrey Paternostro (Fri Jan 15 at 12:00 pm EST)

You shouldn’t bet on baseball.

OK, now that we’ve got that out of the way let’s talk about Wall Street, which some of you probably think of the same way as betting on baseball. One of the sayings you hear about Wall Street is “the market is a discounting instrument.” In other words, the price of a single f stock, like Apple, or a bunch of stocks, like the 30 that make up the Dow Jones Industrial Average, is a reflection of investors’ expectations of the future, discounted back to today. Apple closed at $129.08 on Friday because investors expect it to deliver a stream of earnings and dividends, relative to inflation, to make it worth $129.08 as of 4:00 pm EST on Friday, February 3, 2017.

News can change the outlook. If Apple announces a better-than-expected product launch, investors will increase their expectations of future earnings, and the stock will likely move up. If the government decides to slap a tariff on iPhones made in China, future earnings will be imperiled and the stock will likely move down. I use the word likely because as Joaquin Andujar would say, youneverknow. Some news moves markets, some news doesn’t.

The same is true of betting markets. If there’s a downpour the morning of a horse race, it’ll move the betting markets. Bets on horses known to run well in mud will go up in price and those on horses that prefer a dry surface will go down. If a key player for a football team gets injured during practice before a game, the odds on that game will change. Some news moves markets.

This also applies to baseball betting markets. When a team adds a big star—think the Red Sox signing David Price or trading for Chris Sale—the news moves the odds for the Red Sox. New information should move markets. PECOTA is news. Does it move markets? Let’s run a real-time test.

I’m going to present two tables: One for the American League and one for the National League. I’m going to list, for each team, their current projected PECOTA wins for the upcoming season as well as their odds to win the pennant, per the Superbook sportsbook website. Everything is as of Super Bowl Sunday morning, February 5.

Here are the figures for the American League:

And the National League:

Wait, I know what you’re going to say: “Pennant odds and wins! Apples and oranges!” True enough. Given the divisions in which they play and the way their rosters are constructed, you can make a cogent argument that, for example, the Dodgers are well positioned to win a lot of games but the Cubs are better positioned to win the National League pennant by prevailing in a short postseason series. So if I’m going to compare pennant odds with win projections (which I am), I’m not comparing the same thing.

That’s true, but I don’t think the numbers are going to be all that far removed. The teams that win the most games are generally the ones we favor to win the pennant, and if they’re not the difference shouldn’t be large. We usually don’t expect 85-win teams to be strong pennant contenders or 95-win teams to have no chance. (Additionally, I couldn’t find any over/under figures for wins; they tend to get released later this month. Admittedly, I’m not an expert: I follow the admonition in the first sentence of this piece. If someone’s seen them, please leave a comment.)

I’m going to show you two graphs, one for each league, because I think they illustrate what’s going on better than the 15 rows of numbers in the two tables above. In each graph, the numbers along the x-axis are projected wins per PECOTA. The numbers along the y-axis are the Superbook pennant odds. I used a logarithmic scale on the y-axis, but if you don’t know what that is, don’t worry; it’s not an important detail. Each dot on the graph represents the projected wins and the pennant odds for one team. I labeled each dot so you can see which team it represents. I also included a trend line, for reasons I’ll explain in a minute.

Here’s the American League graph:

To pick out a couple points, "KC" represents the Royals’ 20:1 pennant odds and 70 projected PECOTA wins. "TB" is the Rays’ 50:1 pennant odds and 83 projected PECOTA wins.

Let’s talk about that trend line. It represents the average relationship between Las Vegas pennant odds and PECOTA wins. So the dots above the line—teams whose odds are longer (a higher number) than PECOTA’s projected wins would suggest—are teams that Las Vegas underrates (or PECOTA overrates, depending on your perspective). The dots below the trend line are teams that Las Vegas overrates (or PECOTA underrates).

I found the following relationships interesting:

  • I picked the Royals and Rays as my examples above for a reason. PECOTA likes the Rays and dislikes the Royals (and, for that matter, the Orioles) way more than the betting markets. None of this is new.
  • East coast bias alert! The betting markets like every team in the American League East (other than the Rays) more than PECOTA. As I’ll explain below, this probably illustrates the mindset of gamblers as much or more than that of the odds-makers.
  • Which PECOTA projections are going to be the most controversial? The Royals are, perennially. Orioles too. The Astros and Indians projections, each higher than that of the Red Sox, are probably going to raise the hackles of RSN, particularly given the stat-head-friendly reputations of both this website and the front offices of Houston and Cleveland. (For the record, we don’t give extra credit to analytically-inclined teams.)

Now, the National League:

There’s less divergence between the betting markets and PECOTA in the National League than the American League, probably because the National League has a more contender/pretender, stars/scrubs bifurcation. There’ll be quibbles here and there, but I think there are going to be four major controversies:

  • The Dodgers’ projected win total is nine higher than that of the defending world champions, but the betting markets view them as only tied with the Giants as the third-most likely team to win the pennant.
  • The Cubs are the darling of the betting markets, with 15:4 World Series odds, well ahead of the Red Sox at 11:2 and Indians at 7:1. Anything that doesn’t confirm that is going to be notable.
  • PECOTA likes the Mets more than the Nationals—only one game better, but still—which will raise eyebrows both around the Potomac and among the all-is-lost-if-they-don’t-trade-Duda-and-Lagares-for-Trout faction among Mets fans.
  • Four words: The St. Louis Cardinals. I’m not going to elaborate; I have every confidence that Cardinals fans will let their positions be known here or elsewhere on the site.

If PECOTA moves markets, Las Vegas odds-makers who are BP Premium subscribers are going to log in today, check the 2017 PECOTA projections, and say two words, the first of which is “Oh.” Then they’ll change their numbers, as PECOTA will move the markets.

These appear to be the most likely adjustments, based on the divergence between Superbook and PECOTA:

  • Higher odds for the Cubs, Cardinals, Red Sox, and Royals
  • Lower odds for the Dodgers, Rays, and Twins

But maybe PECOTA won’t move markets. There could be several reasons for this, none of which impugn the PECOTA formulae. (Yes, we hear you, Royals fans, but this isn’t 2015.)

  • The goal of PECOTA is to accurately project the upcoming season. The goal of Las Vegas bookmakers is to make money for their employers. In most cases, those objectives are not in conflict. But when the public is overly enthusiastic about a team (lookin’ at you, Cubs, Red Sox, and Yankees fans), bookmakers can get away with posting less generous odds than they might otherwise, since there will be plenty of gamblers willing to bet on the team no matter what. By that logic, the Cubs are more likely to stay overvalued relative to PECOTA than the Royals, for example.
  • We’re not the first to this game. FanGraphs has had Steamer and ZiPS projections up for a while, giving the betting markets a chance to adjust to analytically-derived standings. (Fun fact: FanGraphs projects 95 wins for the Dodgers, which represents, like our PECOTA projection for the team, the highest win total in the majors. I heard a discussion on MLB Network Radio about how dumb that is, since among other things the announcers insisted, you can’t count on Rich Hill to start 32 games. Steamer has Hill starting 24 games, ZiPS 17, and PECOTA 23.)
  • Reasonable people can disagree. As I said, the sportsbooks want to make money. Hiring dumb people to make dumb projections is not a good way to make money. There are perfectly good reasons for us to expect the Twins to finish just four games shy of .500 while Superbook has them as the least likely team to win the AL pennant. The bookmakers will have confidence in their algorithms, just as we do in PECOTA. They won’t change solely because our numbers differ from theirs.

I’ll check back in a few days and post an update to this article, a BP Unfiltered entry, or, if it warrants, another full article, depending how the odds change.

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
You need to be logged in to comment. Login or Subscribe
You alluded to it in one point but the bookmakers don’t care one bit about who is going to win, they simple work to attract or discourage betting on teams so that no matter wins the house comes out ahead.
If the odds change it won’t be a direct result of PECOTA, but it could be a result of someone (or many someones) with a lot of money betting based on PECOTA.
Good point. I know that for thinly-traded stocks, or horses at small racetracks, the influx of a bunch of money can dramatically move odds. Probably casinos with small sports books, if there's such a thing, would be similarly affected.

Re people betting on PECOTA: I could swear I read a passage to that effect in Joe Peta' very entertaining "Trading Bases: A Story about Wall Street, Gambling, and Baseball," but I couldn't find the reference (or you'd have seen it in the article).
For what it's worth, most Cardinals fans that I've talked to (anecdotal evidence alert) are pretty pessimistic about the team this year.
Yeah, but below-.500 pessimistic? One game better than the Brewers pessimistic? I think that's going to raise eyebrows.
Well, maybe so, but when your expectations are somewhere around .500, it isn't that much of an affront to say your team is going to win 75 games instead. If expectations were that the team was going to compete for the division and have a legitimate shot at the WS and the projection said 75 wins, that would be a problem. But really, the only practical difference between 82 wins and 75 wins is a better draft pick.
I am a Cardinals fan. I was a bit surprised by the 75 wins, but I did not think this was a good team. Curious as to the components that went into the projection, I looked over the player projections, and for the hitters, I did not see any that seemed particularly off. I do think it may, in fact, be a bit optimistic about Peralta.

It seems a bit more pessimistic on some of the starting pitching, particularly Carlos Martinez, than I would be. But overall it is not a great staff. The manager is certainly not a plus, either.

I took my college-aged son to a a couple of Cardinals/Cubs games last year, and I explained to him the young stars we were seeing, the people who he might see with his own son ten years in the future, or might end up in the Hall of Fame. He noted that everyone I pointed out was a Cub. And that is pretty much the Cardinal's problem in a nutshell.
PECOTA, like other projection systems, includes an element of regression to the mean, and when a pitcher like Martinez outperforms his peripherals (3.04 ERA vs. 3.65 FIP and 4.08 DRA in 2016), PECOTA's going to be down on him.

As a kind-of Pirates fan, your assessment of the Cubs is pretty much the whole NL Central's problem in a nutshell, I think.
The above chart results in an average of 80.2 wins in the AL and 78.7 wins in the NL which is much lower than I would expect from simply rounding. Overall 79.5 wins.
I hadn't noticed that. FWIW, since I wrote this, the projections for the Mets and Rockies have improved by a game each, but that pulls the NL average up to only 78.9. Let me look into that...
Mea culpa on this one.

I tune the RS and RA (from which W-L is derived), and it was slightly off this morning. I've adjusted it, so things should be more in order now. Right now, I'm showing 2433-2427 overall W-L, which is about as close as you can get when rounding to a whole number.
As Rob Says, when you're using a run-based projection system, you're more likely than not to have a discrepancy in overall W-L due to rounding. The FanGraphs projections, as of this moment, sum to 2432-2428.

To address bhacking's observation, relative to the tables in this article, add one win to each of HOU, OAK, BOS, TB, BAL, CLE, DET, KC, ATL, PIT, and LAD and two wins to all the other teams. This will not affect the conclusions here, since the trendlines move proportionately and there are no teams with outlier changes.
I'm curious what happened to the Twins projection between Feb 5th and today. They are currently at 80 wins, not 77. Seems like a really big jump
Harold, see the comment directly above yours--bhacking noticed an error in the way we were building up wins when I wrote this on Sunday, and it accounted for a two-win adjustment. As for the other win, I don't know what slight adjustment to playing time might've rippled over to move the Twins projection by a win, but it's nothing major.
I'll take the under on 80 ;-)
First of all, you can't bet "against" any of these Sportsbook pennant odds (unlike win totals where you can bet over OR under), so the books add in a lot of built in juice which projects them from being wrong.

Because of this, they'll also tend to be more wrong below the line than above it because a bettor cannot exploit anything below the line.

10 or 20 years ago, Sportsbooks could and would mostly concern themselves with anticipated batting patterns. That is no longer the case. The reason is that there are legions of very smart, well-financed betters out there (not so much with these season-long bets though) so they have to put out good lines otherwise they will get exploited.

For example, say they anticipate equal action on both sides of a Yankee game where they post a line of Yankees 2-1 odds (-210 on Yankees), but they know that the Yankees should only be, say, 5-3 favorites (around -175). In the old days they could get away with it. Can't anymore. Tons of smart bettors will pound the other side at +190 until the line moves to around where it should have been in the first place. The public becomes irrelevant because the proportion of their money to the total market, including the smart bettors is so small, compared to the old days.

So, yes the market will move based on Pecota numbers, as they already moved when the ZIPS and Steamer numbers came out. It's early so they might not move much until we get closer to the season. But by the time we get close to the season, the opening numbers will have moves tremendously to reflect the projections we see online. That is especially true of the win totals because bettors can bet both sides and the spread is "only" 20 or 30 cents (around 5-7% "juice"). I don't know how much juice or vigorish is built into these pennant lines but I imagine it's a lot (maybe 10% or more). You can figure it out approx by seeing how much return you get on your money if you bet every team and assuming their odds of each team winning or someone else's (like Pecota's) - it doesn't matter that much.
Thanks for the explanation MGL! (I live in a horse racing town and don't go to the track; I'm wholly ignorant.) It sounds like the pennant/World Series bets, given where the lines are set, are an effective means of transferring wealth from bettors to the casinos.