# Team statistics applied to individual players

**URL:** <https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735>\
**Category:** The Game Room\
**Created:** [December 2, 2009, 6:54pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735 "2009-12-02T18:54:28Z")\
**Posts on this page:** 20\
**Page:** 1

<div class="post-metadata">

**Author:** ![Munch](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/munch/32/5281_2.png) [@Munch](https://boards.straightdope.com/u/Munch)\
**Post date:** [December 2, 2009, 6:54pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/1 "2009-12-02T18:54:28Z")

</div>

Few statistics are as hotly contested these days as Wins are for baseball pitchers. Every thread involving post-season awards or post-career awards that involve pitchers seem to devolve into an effort of providing alternative measures to take the place of wins - a team statistic that is being used to measure individual performance.

I’ve asked the question multiple times of why a team statistic is relevant to evaluating individual talent, but have yet to get a reply. Maybe it’s an unfair question. Maybe I’m framing it incorrectly. I don’t know.

Either way, are there any other sports out there that use a team statistic to measure the talent of the individual player?

(And I’m not looking for stats that require involvement from another player. QB rating and completion % rely on receivers being able to catch the ball, baseball assists rely on another player recording an out, soccer/baseball/hockey assists rely on another player scoring, etc. But only Wins AFAIK rely on the entire team.)

---

<div class="post-metadata">

**Author:** ![Tom\_Scud](https://avatars.discourse-cdn.com/v4/letter/t/f08c70/32.png) [@Tom\_Scud](https://boards.straightdope.com/u/Tom_Scud)\
**Post date:** [December 2, 2009, 6:57pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/2 "2009-12-02T18:57:15Z")

</div>

I’ve seen won-loss records for quarterbacks in football, though it tends to be a pretty peripheral stat.

---

<div class="post-metadata">

**Author:** ![Hawkeyeop](https://avatars.discourse-cdn.com/v4/letter/h/958977/32.png) [@Hawkeyeop](https://boards.straightdope.com/u/Hawkeyeop)\
**Post date:** [December 2, 2009, 7:09pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/3 "2009-12-02T19:09:13Z")

</div>

Stats based on team performance when a player is in the game vs when he isn’t? Record in big games?  
Saves?

---

<div class="post-metadata">

**Author:** ![Munch](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/munch/32/5281_2.png) [@Munch](https://boards.straightdope.com/u/Munch)\
**Post date:** [December 2, 2009, 7:11pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/4 "2009-12-02T19:11:41Z")

</div>

> [@Hawkeyeop](#):
>
> Stats based on team performance when a player is in the game vs when he isn’t? Record in big games?  
> Saves?

Saves is close. But you’re already in the game in a save situation, and everything except defense is reliant upon you - your offense doesn’t have to do jack. When a starting pitcher begins the game it’s tied and you have to rely on your offense to score at least one run, as well as the defense to not totally suck it.

---

<div class="post-metadata">

**Author:** ![Tom\_Scud](https://avatars.discourse-cdn.com/v4/letter/t/f08c70/32.png) [@Tom\_Scud](https://boards.straightdope.com/u/Tom_Scud)\
**Post date:** [December 2, 2009, 7:14pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/5 "2009-12-02T19:14:24Z")

</div>

[Plus-Minus](http://en.wikipedia.org/wiki/Plus-minus), mainly in hockey, though it’s been used in basketball as well.

---

<div class="post-metadata">

**Author:** ![Munch](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/munch/32/5281_2.png) [@Munch](https://boards.straightdope.com/u/Munch)\
**Post date:** [December 2, 2009, 7:19pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/6 "2009-12-02T19:19:53Z")

</div>

> [@Tom\_Scud](#):
>
> [Plus-Minus](http://en.wikipedia.org/wiki/Plus-minus), mainly in hockey, though it’s been used in basketball as well.

Very interesting. So it’s applied to everyone on the team equally? It’s not adjusted for playing time - interesting. I \*like \*it.

---

<div class="post-metadata">

**Author:** ![RealityChuck](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/realitychuck/32/195_2.png) [@RealityChuck](https://boards.straightdope.com/u/RealityChuck)\
**Post date:** [December 2, 2009, 7:24pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/7 "2009-12-02T19:24:30Z")

</div>

In hockey, they now credit wins to a goaltender. In addition, both baseball and hockey credit players with shutouts.

Saves in hockey is dependent on how good your team defense is – you get fewer save opportunities with a good defense.

All sports keep track of the wins by the head coach or manager, which is totally dependent on the players. If your team is 0-16, you’re out of a job, no matter how bad your personnel is.

RBIs and Runs scored in baseball are both dependent on the play of the other players on the offense.

As for your question, “why evaluate pitcher on wins” part is due to the fact that, in early baseball, yes, the pitcher was responsible for the win. There were exceptions, but the pitcher controlled the game. That’s less true today, since pitchers aren’t expected to go nine innings – 6 is plenty. That makes it less important than it does in the early days. But even nowadays, great starting pitchers have great won-lost percentages over their careers.

---

<div class="post-metadata">

**Author:** ![Munch](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/munch/32/5281_2.png) [@Munch](https://boards.straightdope.com/u/Munch)\
**Post date:** [December 2, 2009, 7:45pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/8 "2009-12-02T19:45:54Z")

</div>

> [@RealityChuck](#):
>
> in early baseball, yes, the pitcher was responsible for the win. There were exceptions, but the pitcher controlled the game.

I really don’t understand this. Is this Dead Ball Era only, when offense was almost non-existent? Because pitchers still contributed negligibly to offense, which is half the game (and 100% responsible for providing that most important of factors for a Win - a Run).

> [@](#):
>
> But even nowadays, great starting pitchers have great won-lost percentages over their careers.

That is a self-fullfilling prophecy.

---

<div class="post-metadata">

**Author:** ![Jimmy\_Chitwood](https://avatars.discourse-cdn.com/v4/letter/j/f475e1/32.png) [@Jimmy\_Chitwood](https://boards.straightdope.com/u/Jimmy_Chitwood)\
**Post date:** [December 2, 2009, 7:51pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/9 "2009-12-02T19:51:17Z")

</div>

Plus-minus is not really a team stat in basketball, though, since the point is to compare points scored and allowed while the player is on the court vs. points scored and allowed while the player is not. If it’s correctly done it isolates the impact of the individual. That’s not the way the hockey one works, though.

I think the main team stat that gets attributed to individuals is rings. “He’s never won the big one.”

---

<div class="post-metadata">

**Author:** ![Tom\_Scud](https://avatars.discourse-cdn.com/v4/letter/t/f08c70/32.png) [@Tom\_Scud](https://boards.straightdope.com/u/Tom_Scud)\
**Post date:** [December 2, 2009, 7:55pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/10 "2009-12-02T19:55:54Z")

</div>

> [@Jimmy\_Chitwood](#):
>
> Plus-minus is not really a team stat in basketball, though, since the point is to compare points scored and allowed while the player is on the court vs. points scored and allowed while the player is not. If it’s correctly done it isolates the impact of the individual. That’s not the way the hockey one works, though.

I thought that was exactly the way the hockey one worked, except that players in hockey tend to be rotated in and out in groups.

---

<div class="post-metadata">

**Author:** ![Munch](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/munch/32/5281_2.png) [@Munch](https://boards.straightdope.com/u/Munch)\
**Post date:** [December 2, 2009, 8:06pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/11 "2009-12-02T20:06:54Z")

</div>

> [@Jimmy\_Chitwood](#):
>
> I think the main team stat that gets attributed to individuals is rings. “He’s never won the big one.”

Good one.

---

<div class="post-metadata">

**Author:** ![Hawkeyeop](https://avatars.discourse-cdn.com/v4/letter/h/958977/32.png) [@Hawkeyeop](https://boards.straightdope.com/u/Hawkeyeop)\
**Post date:** [December 2, 2009, 8:07pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/12 "2009-12-02T20:07:20Z")

</div>

> [@Munch](#):
>
> I really don’t understand this. Is this Dead Ball Era only, when offense was almost non-existent? Because pitchers still contributed negligibly to offense, which is half the game (and 100% responsible for providing that most important of factors for a Win - a Run).

That and strikeouts were rarer as well, so defense had a bigger part on the outcome than it does now. A pitcher throwing a complete still only controls maybe, 35% of the game’s outcome and even that includes a fair amount of luck. Yeah good pitchers will tend to have better records, but they also tend to have better eras, better warps, better k rates etc. The key is finding the stats that correlate best with winning games, and wins is clearly not one of them.

---

<div class="post-metadata">

**Author:** ![John\_DiFool](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/john_difool/32/19543_2.png) [@John\_DiFool](https://boards.straightdope.com/u/John_DiFool)\
**Post date:** [December 2, 2009, 8:11pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/13 "2009-12-02T20:11:35Z")

</div>

The Cy Young elections of Grienke and Lincecum this year both seem to indicate that the concept of wins the #1 CY qualification is starting to fall by the wayside, and with good reason.

---

<div class="post-metadata">

**Author:** ![Tom\_Scud](https://avatars.discourse-cdn.com/v4/letter/t/f08c70/32.png) [@Tom\_Scud](https://boards.straightdope.com/u/Tom_Scud)\
**Post date:** [December 2, 2009, 8:15pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/14 "2009-12-02T20:15:13Z")

</div>

> [@Munch](#):
>
> Very interesting. So it’s applied to everyone on the team equally? It’s not adjusted for playing time - interesting. I \*like \*it.

The weakness (especially for backups) is that you might be substituted in at times that put you at an unusual advantage or disadvantage. It’s magnified for hockey players, who tend to be substituted in groups; picture a hockey team with two decent wings and two scrubs who decide that it’s better to have all their firepower on the first line with the starting center; the back-up center is going to look like dogmeat even if he’s pretty good because he has to carry a couple of no-hopers. But yeah, an interesting stat if you take those kinds of effects into account.

---

<div class="post-metadata">

**Author:** ![Jimmy\_Chitwood](https://avatars.discourse-cdn.com/v4/letter/j/f475e1/32.png) [@Jimmy\_Chitwood](https://boards.straightdope.com/u/Jimmy_Chitwood)\
**Post date:** [December 2, 2009, 8:18pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/15 "2009-12-02T20:18:25Z")

</div>

> [@Tom\_Scud](#):
>
> I thought that was exactly the way the hockey one worked, except that players in hockey tend to be rotated in and out in groups.

I’m probably not doing a good job explaining. In hockey, as I understand it, +/- is the difference between goals my teams scores when I’m out there and goals the other team scores when I’m out there. What happens when I’m not on the ice doesn’t affect anything (I’m proud of myself for remembering to say ice instead of field; that’s how deep my hockey knowledge runs).

The NBA stat measures that number (net points scored when I’m on court minus points scored by other team during the same time) and then compares that number to my team’s +/- when I’m _not_ out there.

So if you’re on a terrible NHL team, you’re going to have a negative rating pretty much no matter what, and if you’re on a great one, you’re going to have a great one. The basketball stat attempts to correct for this such that a good player with terrible teammates can still have a positive rating even if his team gets outscored by 30 every night, because the minutes when he’s on the court are less negative than the minutes when he isn’t.

---

<div class="post-metadata">

**Author:** ![Tom\_Scud](https://avatars.discourse-cdn.com/v4/letter/t/f08c70/32.png) [@Tom\_Scud](https://boards.straightdope.com/u/Tom_Scud)\
**Post date:** [December 2, 2009, 8:25pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/16 "2009-12-02T20:25:12Z")

</div>

> [@Jimmy\_Chitwood](#):
>
> I’m probably not doing a good job explaining. In hockey, as I understand it, +/- is the difference between goals my teams scores when I’m out there and goals the other team scores when I’m out there. What happens when I’m not on the ice doesn’t affect anything (I’m proud of myself for remembering to say ice instead of field; that’s how deep my hockey knowledge runs).
> 
> The NBA stat measures that number (net points scored when I’m on court minus points scored by other team during the same time) and then compares that number to my team’s +/- when I’m _not_ out there.
> 
> So if you’re on a terrible NHL team, you’re going to have a negative rating pretty much no matter what, and if you’re on a great one, you’re going to have a great one. The basketball stat attempts to correct for this such that a good player with terrible teammates can still have a positive rating even if his team gets outscored by 30 every night, because the minutes when he’s on the court are less negative than the minutes when he isn’t.

Ah, interesting. Not sure if I think the basketball system is better or not - it would kind of suck to be the guy who backs up Michael Jordan or Lebron James under the basketball system you outlined.

And thinking about it more, it would also tend to overrate, say, the only true point guard or true interior defender/scorer on a thin team; if the team struggles when you’re off the floor just because they don’t have a minimally acceptable alternative, it doesn’t mean you’re all that great.

---

<div class="post-metadata">

**Author:** ![Jimmy\_Chitwood](https://avatars.discourse-cdn.com/v4/letter/j/f475e1/32.png) [@Jimmy\_Chitwood](https://boards.straightdope.com/u/Jimmy_Chitwood)\
**Post date:** [December 2, 2009, 8:36pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/17 "2009-12-02T20:36:44Z")

</div>

Yes indeedy, and as you might expect, there’s [more information](http://hoopnumbers.com/allAnalysisView?analysis=RAPM&discussion=True) [on the attempts to adjust for those factors](http://www.82games.com/ilardi1.htm) [than you probably care to read](http://basketballvalue.com/index.php).

---

<div class="post-metadata">

**Author:** ![RealityChuck](https://sea3.discourse-cdn.com/straightdope/user_avatar/boards.straightdope.com/realitychuck/32/195_2.png) [@RealityChuck](https://boards.straightdope.com/u/RealityChuck)\
**Post date:** [December 2, 2009, 8:41pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/18 "2009-12-02T20:41:08Z")

</div>

> [@Munch](#):
>
> I really don’t understand this. Is this Dead Ball Era only, when offense was almost non-existent? Because pitchers still contributed negligibly to offense, which is half the game (and 100% responsible for providing that most important of factors for a Win - a Run).

So? The contributed to the win more than any other player. I don’t care how many runs you score, you still need a pitcher who allows fewer. In the dead ball era, when pitchers were pitching 9 innings routinely, it was their pitching that kept the team in the game. Teams score runs as a matter of course, even in dead ball. If you have the skill to keep the other team from scoring, then that’s something to be desired. It’s just that in the dead ball era, the pitcher was responsible for the entire game.

> [@](#):
>
> That is a self-fullfilling prophecy.

I know. I was trying to think like a sabrematrican.

In any case, no matter how you choose to measure a pitcher’s performance over a career, and no matter how you select your top pitchers, you’ll find your selection won a lot of games (and had a good W-L percentage – especially when compared to how well his team did without him). Over the course of a career, the times the pitcher has a hard-luck loss and the times he has a good-luck win tend to even out, so ultimately the W-L record indicates someone who can pitch well. It’s a crude measurement to be sure (and cruder nowadays), but it is a very good rule of thumb.

---

<div class="post-metadata">

**Author:** ![ElvisL1ves](https://avatars.discourse-cdn.com/v4/letter/e/5daacb/32.png) [@ElvisL1ves](https://boards.straightdope.com/u/ElvisL1ves)\
**Post date:** [December 2, 2009, 8:55pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/19 "2009-12-02T20:55:05Z")

</div>

Have you ever seen a fielder’s glove from the dead-ball era? I wear something much more substantial when I shovel snow. If anybody ever made a play at all with one of those, it was an accomplishment. Quality fielding didn’t matter because it didn’t exist. Defense consisted of pitchers either getting strikeouts or getting hitters to hit the ball weakly enough that a fielder could drop it, pick it up off the ground, and throw it in time. Yes, pitchers quite fairly won and lost games.

With all the errors, recorded and unrecorded, at the time, at least the scorekeepers took pity on pitchers by separating earned from unearned runs.

---

<div class="post-metadata">

**Author:** ![Hawkeyeop](https://avatars.discourse-cdn.com/v4/letter/h/958977/32.png) [@Hawkeyeop](https://boards.straightdope.com/u/Hawkeyeop)\
**Post date:** [December 2, 2009, 8:57pm UTC](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735/20 "2009-12-02T20:57:05Z")

</div>

> [@RealityChuck](#):
>
> so ultimately the W-L record indicates someone who can pitch well. It’s a crude measurement to be sure (and cruder nowadays), but it is a very good rule of thumb.

I don’t really disagree with this. Good pitchers tend to win more games than poor pitchers. So do good 3rd basemen. It doesn’t always even out though.

The hall of fame is a pretty big honor and the line betweening deserving and not can be pretty thin. These players toils in the majors for 10+ years for our enjoyment. Don’t you think they deserve an in depth analysis of whether they deserved to be enshrined or not, and not just some crude measurement that gets it right a decent percentage of the time? At least by the people who vote.

Wins are good. Support neutral wins/losses is better. (it measures how many wins a pitcher should have won giving an average environment) We don’t have to hope it evens out, we can make it so.

[Next page](https://boards.straightdope.com/t/team-statistics-applied-to-individual-players/519735.md?page=2)
