Lucky and Unlucky Teams

by Matthew on Jul 16, 2008 1:23 PM PDT in Seattle Mariners Stats

Back in March I wrote about trying to ascertain a team's true talent level during past events. In a nutshell, you don't want to look at actual win-loss record and even Pythag (looking at runs scored and allowed) is faulty, though a lot better than straight won-loss. Instead, you should look toward BaseRuns which helps tell you how many runs the team should have scored and allowed based on the batter-pitcher outcomes that happened.

(In that post I went on to talk about how tRA can be used to improve further on that, but it's a smaller degree of improvement than using BaseRuns over Pythag and for the purposes of this post, would be far too time intensive to do across all teams.)

So, in that spirit, and inspired by LFoJL I present the methodology behind this ranking of lucky and unlucky teams. I have access to BaseRuns data courtesy of The Hardball Times (up through end of play last Friday) and so I have said true talent level for each team. Taking that, I computed the expected wins and losses over a 162 game season for each team. Then, I simply subtracted that number from each team's current won-loss record, projected out to a 162 game season. Here's an example:

Tampa Bay Rays
Actual record: 55-37
Actual Win%: .598
Projected Record: (.598 * 162) = 96.8 wins
BaseRuns, Runs Scored: 444.81
BaseRuns, Runs Allowed: 364.5
BaseRuns, Expected Win%: (pythag formula) = .592
Projected Record: (.592 * 162) = 95.9 wins
Difference: +0.9 wins (positive = lucky)

There you have it. Pretty simple eh? Alright, time for the rundown. Keep in mind the difference is projected over a 162 game season. It does not reflect the number of games at this current point in time.

FIVE LUCKIEST TEAMS IN BASEBALL

Anaheim Angels. 95.8 (Actual). 78.7 (BaseRuns). +17.1 (Luck)
Minnesota Twins. 90.6. 76.9. +13.7
Pittsburgh Pirates. 75.7. 63.2. +12.5
Cincinnati Reds. 77.6. 69.5. +8.1
St. Louis Cardinals. 89.6. 83.4. +6.2

(Honorable Mention: Astros (+6.0) and Marlins (+5.5) )

FIVE UNLUCKIEST TEAMS IN BASEBALL

Atlanta Braves. 74.9. 89.1. -14.2
San Diego Padres. 64.5. 74.3. -9.8
Cleveland Indians. 68.7. 76.5. -7.8
Colorado Rockies. 67.2. 74.9. -7.7
Seattle Mariners. 62.7. 69.0 -6.3

(Honorable Mention: Blue Jays (-5.9) and Red Sox (-5.2) )

Tweet Comment 69 comments | 3 recs |

Do you like this story?

Comments

FWIW

The Cubs at +1.3 and Brewers at +3.4 round out the NL Central meaning that all six teams have been lucky, to the tune of a combined 37.5 games (over 162).

by Matthew on Jul 16, 2008 1:26 PM PDT reply actions

I know I shouldn't be deathly afraid of a oncoming Brewers

but godammit this is the perfect to time to panic.

by JI on Jul 16, 2008 1:27 PM PDT up reply actions

*a ---> the

by JI on Jul 16, 2008 1:30 PM PDT up reply actions

That was really quick

How are the Red Sox?

by Last Fan Of Jose Lopez on Jul 16, 2008 1:27 PM PDT reply actions

How are the Red Sox what?

by Matthew on Jul 16, 2008 1:31 PM PDT up reply actions

In terms of luck

by Last Fan Of Jose Lopez on Jul 16, 2008 1:31 PM PDT up reply actions

-5.2

see honorable mention under unlucky,

by Matthew on Jul 16, 2008 1:31 PM PDT up reply actions

And they will get Ortiz back.

by JI on Jul 16, 2008 1:32 PM PDT up reply actions

I forgot my reading glasses

Where could I find team BABIP and team LD%? I’m curious to find out the luckiest offenses

by Last Fan Of Jose Lopez on Jul 16, 2008 1:34 PM PDT up reply actions

THT Team page

by Jeff Sullivan on Jul 16, 2008 1:34 PM PDT up reply actions

Cool thanks

by Last Fan Of Jose Lopez on Jul 16, 2008 1:35 PM PDT up reply actions

No Strasburg for you, Padres

regression is gonna be a bitch

The Jose Lopez Watch - 113 H - 12 BB - 67 G Left

by seattlebruin on Jul 16, 2008 1:31 PM PDT reply actions

Any chance we could get the numbers for each team?

by The Beer Baron on Jul 16, 2008 1:33 PM PDT reply actions

Every team not mentioned is +/- 5 games. It's not really interesting.

by Matthew on Jul 16, 2008 1:39 PM PDT up reply actions

Matter of opinion.

I happen to think teams performing as expected are very exciting/fascinating.

by LantermanC on Jul 16, 2008 1:51 PM PDT up reply actions

.

Anaheim Angels. 95.8 (Actual). 78.7 (BaseRuns). +17.1 (Luck)

Would this be a record?

by JI on Jul 16, 2008 1:33 PM PDT reply actions 1 recs

No idea since I've never done this before, but man, that's a ton of luck.

FWIW, they finished 2007 at +9 wins in luck too.

by Matthew on Jul 16, 2008 1:40 PM PDT up reply actions

zomg tha ANGELS hav GOD on their sdie

by Last Fan Of Jose Lopez on Jul 16, 2008 1:41 PM PDT up reply actions

zomg no religion.

JI/Robert '08!

by Fin on Jul 16, 2008 3:12 PM PDT up reply actions

It's all Scioscia .

by JI on Jul 16, 2008 1:41 PM PDT up reply actions

And before anyone says anything

no, it would be silly to suggest that the Angels are doing something systematic to defeat Lady Luck. They’re just lucky.

by Jeff Sullivan on Jul 16, 2008 1:41 PM PDT up reply actions

*It would be silly without proof.

by JI on Jul 16, 2008 1:44 PM PDT up reply actions

Proof!

the other angels fan

by Eyebrows on Jul 16, 2008 3:36 PM PDT up reply actions

Haven't seen you around in a while Eyebrows, where you been?

AKA "BRO-RAY" according to drunk Graham...

by Thingray on Jul 16, 2008 4:20 PM PDT up reply actions

I've been around

Just more time reading/lurking and less time posting. Things are hectic at work lately, so I’m not posting as much.

the other angels fan

by Eyebrows on Jul 16, 2008 4:54 PM PDT up reply actions

I hear you on that one. It's been a hell of a busy month for me.

AKA "BRO-RAY" according to drunk Graham...

by Thingray on Jul 16, 2008 4:56 PM PDT up reply actions

Since 2001 the only Angels tem to underplay it's pythag?

the 2002 Angels.

by JI on Jul 16, 2008 1:49 PM PDT up reply actions

What were we last year?

We could’ve had +7 wins due to our runs scored versus runs against, and then maybe another +7 wins due to getting more runs or preventing more than we deserved.
Just curious.
Also, too bad the Nationals aren’t one of the unluckier teams….

by LantermanC on Jul 16, 2008 1:36 PM PDT reply actions

Do you think Bowden would want Washburn to anchor the staff?

by JI on Jul 16, 2008 1:36 PM PDT up reply actions

He provided a link with this information.

http://www.lookoutlanding.com/2008/3/4/153119/6871

by Jeff Sullivan on Jul 16, 2008 1:36 PM PDT up reply actions

In the original post

According to BaseRuns, the Mariners should have scored 783 runs and allowed 814, a pythag record of 78-84.

by Matthew on Jul 16, 2008 1:37 PM PDT up reply actions

What the heck?

by JI on Jul 16, 2008 3:00 PM PDT reply actions

My fault.

by Matthew on Jul 16, 2008 3:01 PM PDT up reply actions

I'll get over it.

by JI on Jul 16, 2008 3:42 PM PDT up reply actions

Did I offend up there?

If so, sorry.

by Sec 108 on Jul 16, 2008 3:53 PM PDT up reply actions

My guess is that it was LFOJL's usual off-topic hijinks

by Gomez on Jul 16, 2008 3:56 PM PDT up reply actions

You don't need to guess.

by JI on Jul 16, 2008 3:58 PM PDT up reply actions

Well, yeah :P

I just didn’t want 108 thinking he did anything wrong.

by Gomez on Jul 16, 2008 4:02 PM PDT up reply actions

I appreciate it Gomez

I try not to mess with the natural order here since I am not really one of the regulars.

by Sec 108 on Jul 16, 2008 4:03 PM PDT up reply actions

Yeah, I was also playing around with hiding and unhiding stuff

thinking I could hide the first comment of a subthread, thereby hiding all of it and then later go back and unhide that same first comment and unhide the entire subthread.

Turns out it doesn’t work that, if you hide a comment, SBN 2.0 goes through and marks every comment under it as hidden as well and this does not work in reverse. So to unhide the off topic subthread, I’d have to go in and manually unhide every single comment.

by Matthew on Jul 16, 2008 4:22 PM PDT up reply actions

This is why I don't ever close pages

by Last Fan Of Jose Lopez on Jul 16, 2008 4:24 PM PDT up reply actions

FASCIST CENSORSHIP

by JI on Jul 16, 2008 4:33 PM PDT up reply actions

EXPERIMENTAL FASCISM

... wait

by Gomez on Jul 16, 2008 4:34 PM PDT up reply actions

Eh, it's not like what I said was funny anyway

And if it inspired misbehaving it should be hidden.

by Sec 108 on Jul 16, 2008 4:43 PM PDT up reply actions

Fortunately the Rays haven't been too lucky.

I am excited to see them make a second half run to the division.

JI/Robert '08!

by Fin on Jul 16, 2008 3:14 PM PDT reply actions

I'd worry about the Bosox having been unlucky

If they regress, the Rays could be in trouble.

It may have started… they dropped 7 in a row and the Bosox have taken back the AL East lead.

by Gomez on Jul 16, 2008 3:53 PM PDT up reply actions

There's always the WC.

by JI on Jul 16, 2008 3:55 PM PDT up reply actions

They are in the driver's seat

If they were actually good this year, I’d worry about the Yankees.

by Gomez on Jul 16, 2008 3:59 PM PDT up reply actions

I don't think they would be able to duplicate the same second half record they had last year

Especially now that the Rays are good.

JI/Robert '08!

by Fin on Jul 16, 2008 4:08 PM PDT up reply actions

The Yankees have much injury/collapse potential

I’d bet against them.

by JI on Jul 16, 2008 4:18 PM PDT up reply actions

And Joe Girardi will probably end up being blamed for the collapse.

JI/Robert '08!

by Fin on Jul 16, 2008 4:26 PM PDT up reply actions

I blamed Joe Torre for the collapse of the WTC.

I fucking hate you Mariners

by kentroyals5 on Jul 16, 2008 6:25 PM PDT up reply actions

I blame Jeter

any other shortstop would have caught them.

Nice Guys Finish Third - Hopelessly lost, but makin' good time.

by pdb on Jul 16, 2008 9:30 PM PDT up reply actions 1 recs

The one intended for the Capital that crashed in Pennsylvania...

was actually thrown by Chuck Knoblauch.

I fucking hate you Mariners

by kentroyals5 on Jul 16, 2008 9:39 PM PDT up reply actions 1 recs

I wish I could rec this entire subthread.

by Aaron Campeau on Jul 16, 2008 9:43 PM PDT up reply actions

Considering where we are in the Strasburg race

I’d say that we’re pretty damn lucky.

by katal on Jul 16, 2008 4:15 PM PDT reply actions

FAIL.

Incorrect, Matthew.

Read this.

True talent = weighted, regressed sample data (performance).
BsR = sample data (performance)

A team that scores more (or allows less) than indicated by BsR (or win more or less than indicated by BsR or run differential) may indeed be called lucky. But to say that a team’s sample data (performance based on BsR) is their true talent is massively wrong – and I’m getting on your case because you ought to know better! We wouldn’t call any 2008 metric – OPS, tRA, whatever – a player’s true talent without first weighting and regressing.

The Angels are actually a good team – their regressed, weighted sample data indicates they are probably a high-80s wins true talent team. Their performance is that of a high-70s wins team. Their actual winning percentage is that of a mid-90s wins team. All three of those are very different things.

Children, until we have taught them better, will be perfectly happy with a seasonal round of games in which conkers succeeds hopscotch.

by salb918 on Jul 16, 2008 9:03 PM PDT reply actions

This is a great point Sal.

I don’t mean to come across that BaseRuns = team’s true talent level. Especially this early in the season. The intent of this comparison was simply between actual WP and performance to date.

Though I am curious where you get MGL’s numbers from. Scanning the thread you linked to, the only thing I could find was MGL’s explanation, thusly

I simply take my updated pitcher and position player projections, pro-rate them for projected playing time for the rest of the season (ironically, I use BP fantasy charts for that), and "play out" the rest of the season, using each team’s projected wp and a log 5 matchup formula for each game. I play out 10,000 seasons, the post-season, etc., and voila! Nothing fancy. The only trick is to use a decent wp for each team based on their player projections and some approximation of playing time for each player, which includes everyone’s chance of injury, chance of getting called up from the minors, sent down, etc.

But I didn’t see any link to actual regressed weighted samples.

by Matthew on Jul 16, 2008 10:10 PM PDT up reply actions

And even if we assume that these samples exist and are methodologically sound before seeing them....

The Angels are still on pace to play about 10 wins above their “true talent level” win performance, and even under the author’s own conditions that can’t be attributed to anything but luck. So yes, they’ve been lucky and should regress back towards an 87-ish win team rather than the 97-ish win team they’ve played as so far.

by OlSalty on Jul 16, 2008 10:23 PM PDT up reply actions