The Playoff Picture Has Already Shifted

The baseball season, as people will often tell you, is a marathon, not a sprint. It might be even more like an ultra-marathon, one of those races that lasts a hundred miles in the middle of the desert. Once it gets going, the regular season feels like it goes on forever. The idea here is to downplay the significance of any short spurt. There’s an awful lot of time for the standings to normalize. At some point, every team will look very good, and every team will look very bad.

Now, I’ve never run a marathon, myself, but I do know that, among competitors, the splits are critical. When you’re trying to keep up with champion runners, it can be incredibly difficult to make up for a substandard start. That is, even in a marathon, a slow first mile is a problem. A fast first mile can be good or bad — this is where the comparison breaks down. It’s harder for a baseball team to over-tax itself in the first handful of games. But anyway, starts matter. The baseball season has started, and as a consequence of the results, we’ve already seen some numbers shift around.

This is a post I like to get to early in every season. I always get a kick out of it, but at the same time, I also always know it causes people to roll their eyes. It’s way too soon to think about how the season’s going to finish. The season only just began. Games are getting called because it’s too cold, or there’s too much snow. The current leader in position-player WAR is Didi Gregorius. It takes a while for the season to get into a routine, as players are still ramping up to full speed. Mike Trout isn’t yet slugging .550. We all have to be patient.

But this post is a reminder that it’s never too early for the games to count. Every game in the books already has gone for a win or a loss. Every game in September will go for a win or a loss. Of course, the landscape has changed. We have new information. Last year’s Blue Jays and Royals could never fully recover from slow starts. The Rockies took advantage of a hot start. Let’s take a look at how our numbers have already changed.

You know we have projected standings. We have current projected standings, and we have projected standings from the day before the start of the regular season. Here are all the changes in expected final win total.

You Aren't a FanGraphs Member
It looks like you aren't yet a FanGraphs Member (or aren't logged in). We aren't mad, just disappointed.
We get it. You want to read this article. But before we let you get back to it, we'd like to point out a few of the good reasons why you should become a Member.
1. Ad Free viewing! We won't bug you with this ad, or any other.
2. Unlimited articles! Non-Members only get to read 10 free articles a month. Members never get cut off.
3. Dark mode and Classic mode!
4. Custom player page dashboards! Choose the player cards you want, in the order you want them.
5. One-click data exports! Export our projections and leaderboards for your personal projects.
6. Remove the photos on the home page! (Honestly, this doesn't sound so great to us, but some people wanted it, and we like to give our Members what they want.)
7. Even more Steamer projections! We have handedness, percentile, and context neutral projections available for Members only.
8. Get FanGraphs Walk-Off, a customized year end review! Find out exactly how you used FanGraphs this year, and how that compares to other Members. Don't be a victim of FOMO.
9. A weekly mailbag column, exclusively for Members.
10. Help support FanGraphs and our entire staff! Our Members provide us with critical resources to improve the site and deliver new features!
We hope you'll consider a Membership today, for yourself or as a gift! And we realize this has been an awfully long sales pitch, so we've also removed all the other ads in this article. We didn't want to overdo it.

Based on our own math and depth charts, the Mets are already up 4.5 wins. The Angels are right behind them, at +4.2, and then we have the Diamondbacks at +4.0 and the Pirates at +3.6. To round out the top five, the Red Sox are at +2.8. At the other end, the Rays are already down 3.6 wins. The Padres, Dodgers, and Reds are all at -2.6, and then the Orioles are at -2.5. Many teams, of course, haven’t shifted very much, because all the teams have barely played any baseball, but the strong starts and the bad starts all mean something.

It’s arguably more important to look at the playoff odds instead. So here’s the same plot, only this time looking at the changes in the playoff odds, rather than wins.

Viewed like this, no one has had a more encouraging or productive start than the 8-1 Mets. The Mets’ chances of making the playoffs, according to us, have gone up 24 percentage points already. It’s not just that Noah Syndergaard looks good and healthy. It’s not just that Jacob deGrom looks good and healthy. The bullpen has been outstanding, Michael Conforto is back and hitting the crap out of the ball way ahead of schedule, and the club recently wrapped up a three-game road sweep of the Nationals. The Mets are 3.5 games ahead of the Nationals, today. They used to be projected 8.2 games worse. The gap has narrowed considerably. It’s never too soon to sweep your main rival.

The Diamondbacks’ odds are up about 19 percentage points. Similar to the Mets, they’re 8-2, and they recently swept the Dodgers. They also won series against potential wild-card rivals Colorado and St. Louis, so the same ideas apply. The Diamondbacks are in better division and wild-card position, and they’ve already been weathering the absence of Steven Souza Jr. Patrick Corbin looks like a new pitcher. There’s a lot to like down in the desert.

The other big odds jumper would be the Angels, up roughly 18 percentage points. In part, this is because the Angels are 8-3. And in part, this is because Shohei Ohtani has already improved both his hitting and his pitching projections. Ian Kinsler is about to come off the DL, which is further good news. And Andrew Heaney should be back soon. On the downside, the Angels have lost J.C. Ramirez for the year, and Matt Shoemaker is also having arm problems, so it’s not as if everything has been sunshine and rainbows. There are very real questions about the Angels’ pitching depth, but Ohtani has helped to fend them off. In our five-team projected AL playoff pool, the Angels have replaced the Blue Jays.

The biggest drop belongs to the Cardinals, whose odds are down about 11 points. At 4-6, the Cardinals haven’t been dreadful. And now, they have Greg Holland on the roster. But, well, Holland was horrible just yesterday, in his season debut. And this is mostly about the wild-card competition. The Cardinals began as favorites for WC1, clear of the NL’s sixth-best team by 6.3 games. Now the Cardinals project as favorites for WC2, clear of the NL’s sixth-best team by only 1.5 games. The Mets’ hot start matters to them. The Diamondbacks’ hot start matters to them. Even the Pirates’ hot start matters to them. The wild-card races are always tight, and a team can lose ground in a hurry.

A hot streak can erase a bad start very quickly. It works just the same in the other direction. But in any post like this, it’s almost obligatory to mention the gambler’s fallacy. An underachieving team shouldn’t be expected to suddenly overachieve, in order to get back to its regular projection. And, vice versa. Even if you don’t think that, say, the Mets have really improved at all, that doesn’t mean you should now expect them to slump, to negate the eight wins in nine games. While it’s true that baseball is hopelessly cyclical, this isn’t how seasons are understood to work. If you thought the Mets might win 51% of their games before, you should expect them to win about 51% of their games the rest of the way. And then you add on the eight wins and one loss. Teams off to good starts won’t suddenly find their opponents tougher, and teams off to bad starts won’t suddenly find their opponents easier. For most of you, this is old news, but it still always comes up. That’s the thing about popular fallacies.

The purpose here is not to suggest that anything’s been determined. And the purpose here is not to suggest that you should already be thinking about the playoffs, nearly six months ahead of time. The only purpose is to serve as an annual reminder that every game matters. For a team with hopes of contention, every early comeback counts, and every early meltdown stings. It’s never too soon for a game to matter, so it’s never too soon for a fan to emote. That’s you! Consider yourself free to care. That’s the whole point, anyway.





Jeff made Lookout Landing a thing, but he does not still write there about the Mariners. He does write here, sometimes about the Mariners, but usually not.

42 Comments
Oldest
Newest Most Voted
motown odenMember since 2024
8 years ago

Not that they are relevant to anything with “Playoff” in the title, but how does a dreadful Tigers team go 4-5 in their first nine games, and somehow decrease their expected number of wins? The FG projections had them as a .421 team, and they are at .444, but their expected number of wins has decreased. I don’t get it.

HappyFunBallMember since 2019
8 years ago
Reply to  motown oden

It’s more than just a combination of their YTD record with their preseason wins projection prorated for the remaining number of games.

It’s also the changes to DET players’ individual projections and playing time based on results and usage to date. But more importantly is probably the ripple effects of improvements in their opponents. Those extra wins for LAA, BOS, and TOR have to come from somewhere, and it isn’t only going to be division games. DET could be just as good as they were projected two weeks ago, but if their opponents are better that means DET will lose more games.

motown odenMember since 2024
8 years ago
Reply to  HappyFunBall

The point about individual projections, etc., makes total sense, but the article really details w/l record so far. IMO it would be weird to change projections of any hitters too much, they’ve played in maybe one game above 40 degrees, so of course those are going to be tough. Any punishment to hitters should be rewards to pitchers, but who knows.

HappyFunBallMember since 2019
8 years ago
Reply to  motown oden

While 1/16 of a season isn’t much cause to change a projection, it isn’t nothing. But it’s more about playing time when it comes o individual players. Things like Michael Conforto coming back a month early, for example, can significantly affect a team’s projection.

MoateMember since 2022
8 years ago
Reply to  motown oden

Without knowing the specific answer: Opponent quality, injuries to key players, run diff. below expectations, etc.

The team might appear to be “on pace” but it could be that their peripherals show that this is their ceiling, not their baseline, performance.

formerly matt wMember since 2025
8 years ago
Reply to  Moate

Specifically, I bet some of it’s opponent quality. They got swept by the Pirates and then went 4-1 against the Royals and White Sox, before losing to the Indians (don’t know if that’s rolled in to the projections). Since the Royals and White Sox were projected to be really bad and the Pirates were expected to be not good, I’m guessing that even a projected .421 team would be expected to do better than .444 in nine games against those teams.

This doesn’t seems like it should account for two whole wins, though.

Kevin Grimes
8 years ago
Reply to  motown oden

I would guess that because they’re performing more or less to expectation and expectation is that they will not contend, and will trade off assets extrapolating to fewer wins later in the season? If they had started 8-1 or 15-5 or whatever, perhaps that shifts the paradigm?
JUST SPITBALLIN.
WHICH IS AGAINST THE RULES.

Kevbot034
8 years ago

Good stuff. Do you think Braves hot start and surprisingly good OF play holds Acuna back at all? And how do you think that might impact their playoff odds?

smc5784
8 years ago

Thanks for making me feel better about getting emotionally involved in some tight Braves games.

Also, feels like the Braves are a little unfortunate in only being 6-4 with the hot start they’ve had. Had 2 very winnable games that they lost. Not going to bat .400 w/ RISP all year, so I wish they’d made a little more hay in that stretch.

TKDCMember since 2016
8 years ago
Reply to  smc5784

Well, there are more ways you can look at this, right? The Braves are unfortunate in the sense that they have outscored their opponents by a margin that would be indicative of a better record. They have scored a ton of runs. But on the other hand they are fortunate that they have scored all those runs. They are fortunate in two ways. One, their hitters are performing at a much higher level than their true talent (their combined 118 wRC+ would generally coincide with being the best offense in baseball. The Braves are not the best offense in baseball, I’ll turn in my homer card for saying that if necessary). Further, their offensive performance, as good as it has been, has also benefited from good sequencing (as you mention). So they are doubly lucky in that they are hitting much better than their true talent and those hits are coming at the right time to maximize scoring.

But then they are winning games by lots of runs and losing close games (2 of their 4 losses were in extra innings, plus the 2-0 loss last night). You could call that bad luck, but I think if you are a Braves fan you should look on the bright side that you are 6-4 and playing .600 ball is really good.

smc5784
8 years ago
Reply to  TKDC

Good points all around. Obviously, they’ve been very lucky in sequencing and, although I don’t have the number in front of me, I’d have to think with BABIP. What I was trying to say is, I wish they had won a couple more games in the midst of that lucky stretch, since it definitely won’t last.

And, no, keep your homer card. If they end up the best offense in baseball I’ll take a lap around my (very old and conservative) neighborhood naked.

Roger McDowell Hot Foot
8 years ago

“It’s not just that Noah Syndergaard looks good and healthy. It’s not just that Jacob deGrom looks good and healthy.”

In fact, though it hasn’t shown up in the top-line results, they’ve both looked healthy and not all that great so far! While the bullpen will clearly be doing some regression to the mean and the back end of the rotation is still a giant question mark, you’d have to expect that there’s still room for both of those two guys to perform better. Neither one has had a dominant outing yet.

sadtromboneMember since 2020
8 years ago

The Curse of Corey Dickerson

Dieter Von Wolfhausen
8 years ago
Reply to  sadtrombone

Hey there, CoDick ain’t so bad

Free Clay Zavada
8 years ago
Reply to  sadtrombone

The curse of consistently bad front office decisions for several years now

sadtromboneMember since 2020
8 years ago

I agree. It’s looking like other teams are catching up to them, and I don’t think they have any idea where they’re going. They make a ton of smart moves, but they just don’t add up.

FFS, look at their infield. Brad Miller-Joey Wendle-Adeiny Hechevarria-Matt Duffy is depressing. It’s like they’re trying to avoid scoring runs.

Wegandi727
8 years ago
Reply to  sadtrombone

To be fair, it doesn’t make much sense to block Adames and Bauers if you expect them to contribute significantly in the future (this being their expected year to get 100+ games at the MLB level). This years IF was never expected to stick around long. Adames, Bauers, and Arroyo are all Top 100 IF prospects. While you don’t expect to hit on all three blocking playing time isn’t smart. Miller is more of a DH anyways (this is a sad sentence though) at this point. Duffy is a respectable player to have in the IF. Just because he has been hurt the past couple years doesn’t diminish what he was prior to that. A healthy Duffy is a good asset.

Mike NMN
8 years ago

Well, I guess there goes the rest of my Yankee-rooting season. That being said, wins are wins and losses are losses and once they are in the book, they aren’t going to be changed. The interesting question with these types of analysis is are context. If these ten games occurred between game 60 and 69, would they be noteworthy?

bananas
8 years ago
Reply to  Mike NMN

The yankees odds haven’t changed significantly at all….?

tb.25
8 years ago
Reply to  Mike NMN

If these ten games occurred between game 60 and 69, would they be noteworthy?

Yes. Every 10 game set has significance. A win is a win.

the wiener
8 years ago

I feel like this statement ‘ Teams off to good starts won’t suddenly find their opponents tougher, and teams off to bad starts won’t suddenly find their opponents easier. ‘ contradicts the rest of the article, in that you are saying that teams off to hot start should expect a higher win total at the end of the year. So that means they are better team than expected, so team playing them should expect to suddenly find them tougher, and therefore while teams that are off to hot starts wont suddenly find all opponents tougher they will find some of them tougher than expected, and some of them easier, based on their changed win total projections.

stever20Member since 2017
8 years ago
Reply to  the wiener

what he’s saying is that a team that was projected to go 81-81, if they start off 8-2, you can’t just assume the rest of the way they’re going to go 73-79- to make up for that start. To regress to what was originally expected.

Last year Colorado wasn’t projected to do much- and started off 33-22. Don’t think many expected them to be more than around a .500 team. Rest of the season they played at a 54-53 pace.

Roger McDowell Hot Foot
8 years ago
Reply to  the wiener

I love how every time an article here gives a paragraph of exposition about the gambler’s fallacy, someone shows up to argue against it in the comments.

sadtromboneMember since 2020
8 years ago

To be fair, it is not an easy concept to understand. It requires a working knowledge that (a) the underlying probability distribution of “winning a game” hasn’t changed that much but (b) that what has already happened is in the books and can’t be changed. Push too hard on one point and you lose people on the other.

Roger McDowell Hot Foot
8 years ago
Reply to  sadtrombone

It’s clear based just on this phenomenon itself that it’s not an easy concept to understand! I was trying to think what else gets this kind of treatment from the Internet and the only thing that occurred to me was the Monty Hall problem.

sadtromboneMember since 2020
8 years ago

I just googled the Monty Hall problem (being only loosely familiar with it), and that is way harder to understand than the gambler’s fallacy. Heck, I don’t even understand it. Although from what I’m reading simulations show that you should switch doors, I don’t understand why.

shortstop
8 years ago
Reply to  sadtrombone

Let’s say you pick door #1 initially. There is a 1/3 chance door #1 is correct, meaning there is a 2/3 chance door #2 or door #3 is correct. It is revealed that door #2 is not correct, but the fact remains that p(door #2) + p(door #3) = 2/3. Therefore, if p(door #2) = 0, then p(door #3) = 2/3, while p(door #1) still = 1/3.

Roger McDowell Hot Foot
8 years ago
Reply to  sadtrombone

It’s a lot more strongly counterintuitive, for sure (at least if your intuition runs the way most people’s does — the important thing to realize about it is that it’s about how the new information you receive midway through changes the problem). But I just meant it was the only other case I could think of where people will not just hear the explanation, but really consistently show up to argue at length that their intuition is right, and probability/statistics is wrong.

The Kudzu KidMember since 2016
8 years ago

You should read some of the stuff people write about Cantor’s diagonal proof that the real numbers are uncountable. Any time good math gives a counterintuitive result people jump to argue against it.

mvillalo
8 years ago

How can LAA, BOS and TOR all have increased playoff odds? Aren’t they competing for the same two spots?

HOU and NYY don’t seem to have lost much/any

TerenceMember since 2016
8 years ago
Reply to  mvillalo

Oakland, Texas, Minnesota, Tampa Bay, Seattle, Baltimore, and Cleveland are seven of the fourteen teams that have significantly decreased. They’re competing against those teams for wild card positions. I would assume the losses of those seven is about equal to the gains of your three, with HOU, CHW, KCR, DET, and NYY being about a wash.

Paul-SF
8 years ago
Reply to  mvillalo

Everyone starts out with at least *some* non-zero chance of making the playoffs. When a bad team starts out poorly – or even starts out OK but the best teams start out hot – those odds diminish further, and the hot teams’ rise accordingly.

Or, shorter answer: Look at all the AL teams who weren’t expected to be particularly competitive and yet have lost ground in the odds: Texas, Tampa, Baltimore, Seattle, Detroit, Oakland. The Angels, Jays and Sox are divvying up those teams’ playoff odds among them.

stever20Member since 2017
8 years ago
Reply to  mvillalo

Twins have gotten out a bit slowly which is something. And yeah, same for Yankees.

WoundedSprinterMember since 2018
8 years ago
Reply to  mvillalo

LAA is competing with the Astros. (Gamely, and I salute that effort.)
BOS and TOR are competing with NYY, and I’d call that a 2 1.2 way contest.
There are two WC spots, which are independent of division.
Predictions are a bit of a juggling thing, but I think we can assume that the basis is “top five teams in the league.” The split between division winner and WC is the bit that probably moves around a lot.

DavidMember since 2016
8 years ago

Cardinals’ opponents are 16-4 vs. other teams so far in 2018, so their 4-6 record is actually .200 better than the rest of the pack… If they can maintain that mark they’ll win well over 100 games. 🙂 (edited for bad math)

tdouglas
8 years ago
Reply to  David

This math makes no sense. They’ll be hard pressed to win 90 with how they’ve begun the season. In the rotation, Wainwright looks toast, and Wacha has been awful (his fastball is down ~2.5mph, which is a MAJOR red flag, especially for a guy with major shoulder issues). In the bullpen, there’s only one decent lefty and a bunch of average righties. Holland helps, but he’s not an elite arm anymore.

There were a ton of question marks in St. Louis going into 2018, and their bad start just exacerbates their issues.

tb.25
8 years ago

I’m a bit surprised the Giants are below expectations.

Having all three of their series against playoff teams from 2017, including two series against the reigning NL champs… I’d have projected a slightly below .500 record. Which they have.

Edit: I understand drops in batter/pitcher projections. But having faced no 4th or 5th starters yet… I think that quality of competition should have had a more significant projection impact.

francis_soyer
8 years ago
Reply to  tb.25

They have a .500 record and if they can maintain this until Bumgarner comes back, they should be better than 50% to make the playoffs.

bananas
8 years ago

I always love this article.

It’s worth noting that the projections page also always includes this data and is fun to explore!

I personally like to go one layer deeper than just “playoff odds” and look at DIV%.

For example you can now see that the red sox have overtaken the yankees to win the East (….it’s still a complete toss-up, but the roles have flipped).

https://www.fangraphs.com/standings/playoff-odds-graphs?lg=AL&div=E&stat=div&year=2018

phealy48
8 years ago

Trout’s slugging .550 now 🙂

Six Ten
8 years ago

I understand statistically how it can happen, but it is still really counterintuitive that zero teams in the AL Central have increased their playoff odds while Cleveland’s and Minnesota’s have dropped off significantly.

stever20Member since 2017
8 years ago
Reply to  Six Ten

not really. Clevelend still a massive favorite for division and the Twins 2nd. The decrease in the Twins/Indians playoff odds got taken up by the Angels/Red Sox/Blue Jays.

Jesse
8 years ago

This article is only a day old and now Trout is slugging over 500. I’ll assume he read Jeff and fell in line.

isavage
8 years ago

I do roll my eyes at these articles. I guess what I’d mostly be curious about, is what are these numbers based on? Is there actually statistical analysis that has been done that shows the first 10 games of the season have significant meaning like this? E.g. if you look at preseason projections, vs. adjusted projections 10 games into the season that you’re using, at the end of the year are the 10 games in projections actually better off at predicting the full season record? Or would you have been better off ignoring the small sample and just going with the preseason numbers?

Logically, I think it’s incorrect to invoke the gambler’s fallacy, because baseball games are not games of random chance. When you have a team projection, built of those individual projections, it’s involving players whose performance varies greatly throughout the season to get to that overall end point. So if a team struggles in the first week because some good players had poor starts, is that actually meaningful in any way? If you have a guy who’s hit .300 with 30 HRs for the past 3 years and he’s hitting .200 with 1 HR in the first 10 games … I would completely ignore that small sample and still expect the guy to hit around .300 with 30 HRs. It’s not gambler’s fallacy to think that, it’s regression to the mean. And the team performance vs projection are made up of all of those individual small samples.

Obviously if someone goes 1-9 to start the season who was supposed to be good, that’s a major change. But I really think dropping someone like Cleveland 2 games because they were .500 10 games into the season while the entire team is hitting below .200 is not based on any reality. They win the next 2 games and now have a better winning % than they were expected to have preseason.