Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Baseball Newsstand > Baseball Primer Newsblog > Discussion
Baseball Primer Newsblog
— The Best News Links from the Baseball Newsstand

Friday, February 21, 2020

A Quick Look at Our Playoff Odds

With the release of full ZiPS projections, our playoff odds are up and running. For the most part that means putting a number to things that we already know. The Dodgers are 97.7% likely to make the playoffs, which sounds about right. The NL Central is a four-way tossup with the Cubs out in the lead. The NL East has three teams each with around a one-in-three chance at it. That all tracks with intuition.

Indeed, for the most part, the standings are self-explanatory. That doesn’t mean that everything is obvious and intuitive, however. Let’s take a quick look at a few of the cases where a deeper dive is necessary.

It’s tempting to think of a team’s expected win total as just a sum of their WAR. After all, the W is right there in the acronym! As Dan notes every year, however, adding up WAR totals on a depth chart isn’t a great way to go about things. Rather than just do that blindly, however, we can look at teams whose projected wins diverge the most from their WAR.

To do that, we’ll need each team’s projected WAR totals. Thankfully, there’s a handy page that shows all that data. The Dodgers have the most projected WAR and the Orioles have the least.

Well, no better time of year to start making such predictions than the present…...

 

QLE Posted: February 21, 2020 at 01:18 AM | 18 comment(s) Login to Bookmark
  Tags: odds

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. villageidiom Posted: February 21, 2020 at 09:38 AM (#5925816)
The Angels have a few problems. First, their schedule is sneakily difficult. They get the Dodgers as a “bonus” interleague team, which stings. Their extra games (not every team in a given division plays an equal out-of-division schedule) relative to the Rangers are against the Rays, Yankees, and Twins. Overall, they just got unlucky on the schedule.
HOW WILL THE ANGELS LOSE THIS SEASON?
   2. crict Posted: February 21, 2020 at 09:48 AM (#5925818)
Impossible to know without good knowledge of the cheating schemes. That's where baseball is now. (I know, still exaggeration).
   3. puck Posted: February 21, 2020 at 10:55 AM (#5925840)
It's amazing that 3 teams project to 98 wins. This seems like a different method of doing the projection, but even if it is WAR-based, that's a high projection.

The Rockies meanwhile have a lot of interpolating to do.
   4. JJ1986 Posted: February 21, 2020 at 11:01 AM (#5925841)
The Mets are somehow tied for the second best team in the NL.
   5. Pops Freshenmeyer Posted: February 21, 2020 at 11:13 AM (#5925843)
It's amazing that 3 teams project to 98 wins. This seems like a different method of doing the projection, but even if it is WAR-based, that's a high projection.

That jumped out at me too. But then I look at New PECOTA's projected standings and it's the same story. In both the Dodgers are projected to win 10+ games more than the second best NL team with the Yankees and Astros over 96 wins.

I think there is good reason to believe that last year was no aberration and MLB is evolving to have more teams on the W/L extremes.
   6. Sunday silence Posted: February 21, 2020 at 11:23 AM (#5925848)
It’s tempting to think of a team’s expected win total as just a sum of their WAR....As Dan notes every year, however, adding up WAR totals on a depth chart isn’t a great way to go about things.


why do they keep repeating this, if so what the point of WAR? I mean something's wrong if WAR doesnt come somewhere close.

To me the only modifier is how much team X is going to spend on trades at the deadline. PIT's WAR is probably not going to change much at the deadline; but NYY or LAD? If projected WAR + whatever spending on WAR teams did at the deadline doesnt come close to w/l, then somethings wroong.
   7. Pops Freshenmeyer Posted: February 21, 2020 at 11:47 AM (#5925860)
why do they keep repeating this, if so what the point of WAR? I mean something's wrong if WAR doesnt come somewhere close.

In this case he's pointing out that team schedules have variable difficulty.
   8. Ron J Posted: February 21, 2020 at 12:04 PM (#5925867)
#7 Sure. And it's more than just the teams you play. Sometimes you catch the Angels with Mike Trout on the DL. Or you get a disproportionate number of 5th (or worse) starters.

I remember looking at this in the past and the difference in real strength of opposition can easily be the difference in a close playoff race.
   9. Rally Posted: February 21, 2020 at 12:14 PM (#5925873)
#7 Sure. And it's more than just the teams you play. Sometimes you catch the Angels with Mike Trout on the DL. Or you get a disproportionate number of 5th (or worse) starters.


That certainly affects strength of schedule, but its not something that can be predicted. We do know how many times each time will play every other team, at least for the first 162 games, so that is not too hard to adjust for.
   10. Stevey Posted: February 21, 2020 at 01:02 PM (#5925894)
So yes, counting up the WAR doesn’t work exactly perfectly. It works pretty well though — there’s a 96% r-squared between WAR-based win estimates and our Depth Charts estimates. And the biggest differences are small in the grand scheme of things. Three or four wins tops, accounting for strength of schedule and sequencing and everything? It doesn’t sound so bad.
   11. Karl from NY Posted: February 21, 2020 at 02:36 PM (#5925915)
In the middle of the article is also a wordy description that boils down to this: Team wins isn't straight addition of player WAR because WAR is context-neutral but a team can create its own context.

Put your OBP guys in the lineup ahead of your SLG guys, and they can take advantage of each other in real context more than WAR would expect in treating every event as neutral context. Conversely, you might have a SLG heavy lineup that's less than the sum of its parts because each one of them has less runners to drive in than WAR would context-neutrally expect.

This principle also applies to pitchers, particularly bullpens. The Mets total WAR comes out to less team wins than expected, because they have several good relievers but they can't all get the high-leverage innings.
   12. Karl from NY Posted: February 21, 2020 at 02:47 PM (#5925920)
The Dodgers are 97.7% likely to make the playoffs

Is this sort of number back-tested at all? If you look back at the history of all teams that begain with a 97% projected chance, or even rerun it for older years using preseason data, do you actually see 97% of them in the playoffs after the fact?

(You could adjust this to account for changing playoff rules - you can make 1995 compatible with 2012 if you say "#5 seed or better" rather than actual playoff entrants)
   13. snapper (history's 42nd greatest monster) Posted: February 21, 2020 at 03:00 PM (#5925924)

Is this sort of number back-tested at all? If you look back at the history of all teams that begain with a 97% projected chance, or even rerun it for older years using preseason data, do you actually see 97% of them in the playoffs after the fact?


I would think we don't have nearly enough seasons with projections to test it. How many years back do projections go? 15? There may only be 8 97% teams in there.
   14. Tom Nawrocki Posted: February 21, 2020 at 03:20 PM (#5925930)
In the middle of the article is also a wordy description that boils down to this: Team wins isn't straight addition of player WAR because WAR is context-neutral but a team can create its own context.


Or, it might be that team wins isn't straight addition of player WAR because WAR isn't particularly accurate or precise and shouldn't be taken nearly as seriously as it is.
   15. Karl from NY Posted: February 21, 2020 at 03:52 PM (#5925937)
I would think we don't have nearly enough seasons with projections to test it. How many years back do projections go? 15?

You don't need to actually have made the projections back then. You can make them now. Take the 2020 calculation model and run it with 1995's data to make your projections for 1996, then compare with the actual results of 1996. This is how ideas for stock picking are back-tested.
   16. snapper (history's 42nd greatest monster) Posted: February 21, 2020 at 04:08 PM (#5925939)
You don't need to actually have made the projections back then. You can make them now. Take the 2020 calculation model and run it with 1995's data to make your projections for 1996, then compare with the actual results of 1996.

Well, until you get out of sample. Today's models will not predict 1935 very well :-)
   17. Walt Davis Posted: February 21, 2020 at 07:24 PM (#5925961)
A team-win projection should be taken as nothing more than (at best):

"Given this assumed distribution of playing time and assuming other teams' playing time assumptions are correct and assuming each player's projected production rate is correct, this is the best estimate of the number of expected wins. Of course we know those assumed playing times will not hold exactly but we can also consider this projection as a reasonable approximation of the number of wins as long as the actual PT and performance distributions aren't substantially different than the assumed ones, keeping in mind that one player's random over-performance may be balanced by a teammate's random under-performance. Injuries, massive slumps, massive hot streaks, prospects unexpectedly breaking out or struggling will also affect the actual PT distribution and of course we can't really model that so it is best considered as a source of uncertainty (variance) rather than bias. There's a lot of uncertainty and our simulations provide some estimate of that uncertainty and it would be nice if we made that so easy to find that even Walt could find it but it's not a perfect world so Walt guesstimates that is at least +/- 8 wins.

"We will further add that anybody who thinks it is possible to accurately project playing time over a 162-game season has no idea what the #### they are talking about. If you aren't clever enough to know that any projection for the Angels is likely to be wrong if Trout misses the entire 2020 season, that's on you.

Love & kisses,
Fangraphs"

However I have only now discovered that they offer 3-year ZiPS player projections. Cool.

Mookie
2020 379 wOBA (#10), 0 fld, 2 bsr, 6.1 WAR (#3)
2021 379 wOBA (#13), 0 fld, 1.5 bsr, 5.8 WAR (#5)
2022 379 wOBA (#13), 0 fld, 1.3 bsr, 5.6 (#6)

Some of those numbers don't quite make sense. In 2022, Bellinger is projected to a much better 397 wOBA, better fielding and even slightly better baserunning but gets a smidgen fewer WAR (in 20 fewer PA). I thought maybe Mookie was being projected as a CF but he's listed under the RF tab so I don't know what's going on. (Both are projected as Dodgers, it's not a stadium thing.) Ahh, for some reason Bellinger is projected as a 1B not a RF and I guess the position adjustment isn't included in the "fld" component but obviously feeds into WAR. Still surprised that Mookie would project as an average RF already (even with heavy regression for defense).

Some slight surprises: ZiPS loves Albies going forward, really likes Bo Bichette, doesn't think Semien was a fluke, believes in Vlad Jr's bat (382 wOBA by 2022). Alas it expects Bryant to stay at 2019 levels. It's decided that Stanton is done as an elite player (2.9 WAR in 2020 even in 567 PA, then 2.4 and 1.9).

Anyway, all of this (well, not the bit I "quoted") can be found on fangraph's projection page
   18. Walt Davis Posted: February 21, 2020 at 08:03 PM (#5925964)
On the non-Dodgers balance in the NL -- it was pretty balanced last year as the NL teams have, to this point, not quite taken to tanking to the degree the AL has. But they have taken to "golly, we're close to the CBT threshold, let's not do anything." The Cubs, Brewers and Cards all pretty well stayed put while the Reds are trying which has helped close the gap. The Braves lost Donaldson, the Nats lost Rendon, the Mets lost Wheeler while adding Betances, Porcello and Wacha. The Phils actually acted a bit like a ML team (Wheeler, Didi, various small moves but non-tendered Hernandez) and even the Marlins acted more like an ML team than the top of the NL Central. The Padres reportedly were interested in various big names but ended up waiting for the kids to break out (they have plenty).

Meanwhile the Dodgers have so much talent they were trying to give away Pederson and Stripling ... and failed. You could split them in two and they might have a decent shot at the playoffs in the NL.

You must be Registered and Logged In to post comments.

 

 

<< Back to main

News

All News | Prime News

Old-School Newsstand


BBTF Partner

Dynasty League Baseball

Support BBTF

donate

Thanks to
Shooty would run in but these bone spurs hurt!
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Hot Topics

NewsblogBaseball in Arizona: Could it Work?
(11 - 7:22pm, Apr 09)
Last: Walt Davis

NewsblogOT – NBA CoronaThread 2020
(85 - 7:17pm, Apr 09)
Last: If on a winter's night a traveling violation

NewsblogTwo main reasons why Giants, fans got iconic Oracle Park design right
(5 - 7:15pm, Apr 09)
Last: SoSH U at work

NewsblogBaseball Question of the Day: Have you ever permanently changed your rooting interests?
(45 - 7:14pm, Apr 09)
Last: phredbird

NewsblogUS-based pro sports leagues monitoring coronavirus outbreak
(4893 - 6:58pm, Apr 09)
Last: cercopithecus aethiops

NewsblogIf 2020 season is canceled, which players would be hurt the most?
(25 - 6:41pm, Apr 09)
Last: The Mighty Quintana

NewsblogAn Update on the State of FanGraphs
(2 - 6:39pm, Apr 09)
Last: McCoy

NewsblogTexas Rangers Hall of Famer Josh Hamilton indicted on charge of injury to a child
(7 - 6:02pm, Apr 09)
Last: Joe Bivens, Elderly Northeastern Jew

NewsblogDo Big-Boned Position Players Age Worse Than Everyone Else?
(15 - 6:01pm, Apr 09)
Last: Joe Bivens, Elderly Northeastern Jew

NewsblogPosnanski: Baseball 100 Rules
(1323 - 5:25pm, Apr 09)
Last: Mefisto

NewsblogDorktown: A History of the Seattle Mariners
(13 - 2:22pm, Apr 09)
Last: villageidiom

NewsblogOT Soccer Thread - Spring 2020
(187 - 1:08pm, Apr 09)
Last: puck

NewsblogTaiwanese baseball team replacing fans with 'robot mannequins' to open season
(12 - 12:35pm, Apr 09)
Last: Gonfalon Bubble

NewsblogEx-MLB player finishes medical school, primes to fight virus
(6 - 11:58am, Apr 09)
Last: Blastin

NewsblogHome Run Derby, Anyone? Eight Ideas to Shorten Extra Innings
(20 - 11:38am, Apr 09)
Last: Pat Rapper's Delight (as quoted on MLB Network)

Page rendered in 0.3531 seconds
46 querie(s) executed