Run Differential

by WittyUserName on Aug 29, 2008 10:07 AM CDT Comment 13 comments

I've been following the standings on MLB.com and have been quite intrigued by the run differential for teams, and the subsequent expected WIN-LOSS records for those teams. Strangely enough, some of teams with amazing records shouldn't, statistically speaking, be having those kinds of years

Here's the breakdown of the top teams in reference to run differential and expected WIN-LOSS:

1. Cubs +193 85-49 (Actual Record: 84-50, -1)

2. Red Sox +122 78-55 (Actual Record: 77-56, -1)

3. White Sox +108 77-56 (Actual Record: 76-57, -1)

4. Phillies +88 76-58 (Actual Record: 73-61, -3)

5. Rays +80 75-57 (Actual Record: 81-51, +6)

6. Mets +73 74-60 (Actual Record: 74-60)

7. Brewers +72 74-59 (Actual Record: 77-56, +3)

8. Jays +66 74-59 (Actual Record: 68-65, -6)

9. Twins +65 73-61 (Actual Record: 75-59, +2)

10. Angels +57 72-61 (Actual Record: 81-52, +9)

11. Cardinals +55 72-62 (Actual Record: 74-60, +2)

Now, obviously having a winning expected record guarantees nothing in reality. However, for the most part the expected records are really close to teams actual records, the only primary exceptions are the Jays, Rays, and Angels

Last season the D-Backs defied logic by winning despite a poor expected record and poor run differential, so having the Angels and Rays win wouldn't be such a shock, and by all accounts, they'd be contenders even if they weren't outplaying their differential. But what does that say about the teams? Obviously it shows that they are winning close games and losing some blowouts, but is there anything else to be read into this?

Does anybody else find these numbers interesting? Based on this, the Red Sox should have a 1.5 game lead over the Rays and a 3 game lead over the Jays. And the Phillies would lead the Mets by 2, as opposed to them trailing by 1. The Angels are clearly winning a ton of close games, and there is clear evidence to that as K-Rod is chasing the single-season saves record

Do people think this is telling of how teams might fair down the stretch, or how they'll perform in October? I think it tells the story of just how dominating the Cubs have been, and how I don't think they need to cower in fear of any opponent. I think it will be interesting to see if these numbers take their toll down the stretch, particularly in regard to the Rays, as the Angels have all but secured themselves that division.

This is a FanPost and does not necessarily reflect the views of SB Nation or Al Yellon, managing editor (unless it's a FanPost posted by Al). FanPost opinions are valued expressions of opinion by passionate and knowledgeable baseball fans.

0 recs | Comment 13 comments

Story-email Email Printer Print

Comments

It's the team's Pythagorean win expectation.

MLB.com uses the following formula:

RS1.82/((RS1.82)+(RA^1.82))

Which is fine for something quick and dirty, but for better accuracy you’d use custom exponents instead of the 1.82.

Yeah, when you look at a team that’s significantly exceeding their win expectation, you figure there’s a possibility of luck being involved. Over a 162 game schedule, there’s a standard deviation of six wins between a team’s true talent level and their actual win-loss record.

BP has their adjusted standings, which take into account strength of schedule as well.

I has a blog. | Ask a sabermagician!

by cwyers on Aug 29, 2008 10:17 AM CDT reply actions 0 recs

Checked those out

From a quick look, it appears that most of the records would be similar to what I posted above.

I know that it is a very quick an dirty way of doing it, I didn’t know how to go about the more in-depth way, which you posted.

Even in those adjusted numbers the Angels shouldn’t be as good as they are, and the Jays should be contending. In fact, the Angels actually get worse as the analysis gets more sophisticated.

by WittyUserName on Aug 29, 2008 10:53 AM CDT up reply actions 0 recs

strength of schedule

i talked about that all last year how people were misleading the run differential between the Cubs and DBacks. The Cubs last year had a negative run differential against teams outside the NL Central and last year the Central wasn’t a very good division.

The one thing that run differential doesn’t do a particularly great job at is comparing teams across leagues. A +100 RD in the NL is not equivalent to a +100 RD in the AL.

RD is a good stat to use and measure between teams in the same division and for the most part teams in the same league as long as there aren’t huge discrepancies in the dispersion of talent within divisions. Unfortunately the unbalanced schedules haven’t helped the RD argument which is why you need to take a look at the BP adjusted standings Cwyers noted

by DartmouthCubsFan on Aug 29, 2008 11:28 AM CDT up reply actions 0 recs

Frequently

teams that are overplaying their expected record have a good W/L record in close games.

This year in 1-run games:
Rays: 25-15
Angels: 25-18

Conversely,
Jays: 19-27

by rlpete on Aug 29, 2008 10:21 AM CDT reply actions 0 recs

Brew Crew as well

25-11 in 1-run games

by chromate on Aug 29, 2008 2:20 PM CDT up reply actions 0 recs

2004 Team

The 2004 team had a run differential of +124. They finished 89-73. Their Pythagorean win/loss was 94-68. 94-68 would have been good enough to win the wild card. The Cubs were only 19-30 in one-run games that season. An injured Joe Borowski and ineffective LaTroy Hawkins contributed to that one-run game record.

"The big possum walks late." - Harry Caray

by memphiscub on Aug 29, 2008 12:18 PM CDT reply actions 0 recs

What about the 2003 team?

I seem to recall a lot of close games.

by dr stabbingworth on Aug 29, 2008 12:47 PM CDT up reply actions 0 recs

2003 Team

The 2003 team had a run differential of +41. They were actually 88-74. Their Pythagorean win/loss was only 85-77. The Cubs were 27-17 in one-run games. 2003 was Borowski’s best season in the big leagues.

"The big possum walks late." - Harry Caray

by memphiscub on Aug 29, 2008 1:06 PM CDT up reply actions 0 recs

Interesting

Thanks

by dr stabbingworth on Aug 29, 2008 3:54 PM CDT up reply actions 0 recs

I love statistics but they are merely indicators and measurements reflecting the bottom line

This is true in almost any endeavor; business, politics, science et cetera. The bottom line is actual wins and losses. The Cubs are 35 games over .500. Digging into the data there are two enormous indicators as to why they are this far above the league——Soriano’s effect and Harden’s addition.

Cubs are: 62-28 with Soriano and 23-22 without Soriano or with Soriano an astounding .688 winning percentage or almost winning 7/10.

Now with Harden Cubs are: 28-13 with Harden, they were 57-37 without him. Again with Harden the Cubs are playing at .682 and without him .606

Now combined since Soriano’s return with Harden the Cubs are 27-9 or playing at .710 ball which is just above 7 out 10. Before Harden joined the club when Soriano played, 34-19 (.641).

You could say there is a small Samardzija factor but essentially he replaced an ineffective Howry role not a difference maker.

Essentially the combination of Soriano to the everyday lineup which makes everyone better to the tune of over 1.4 games over every ten or 6 games over 50 or 18 over a season! Adding Harden to a Soriano lineup is like adding 6 games over 100 and 9 games over a season——if both played the entire season 27 more wins than if the Cubs didn’t have them.

You might be gaga over Ramirez’s RBI stats as MVP or DeRosa’s career season and ability to shore up the injuries but the MVP is Soriano when it comes to wins and losses.

Piniella: "This is a tougher job than I thought it would be, I'm going to be honest with you."

by Ivy Walls on Aug 30, 2008 10:07 AM CDT reply actions 0 recs

+1

I’ve been saying for while that Soriano deserves MVP consideration. Sure he makes some boneheaded plays from time to time, but you can’t argue the record with him verses without him. Most of the voters would say he missed too many game to be considered, but those games he missed proved his value. If the team is playing .500 ball without him and .688 ball with him doesn’t prove his value?

"Destiny is a matter of choice, not chance"

by MerlinDog on Aug 30, 2008 11:14 AM CDT up reply actions 0 recs

Agree about Sori (sorta), disagree (sorta) about Harden

Soriano
Soriano has been great, and the difference between the team with him and without him is huge as documented above. But you cant quite say he is the MVP based on that. What would the teams record without Aramis? We dont know, because he has stayed healthy. I would argue that the fact he stayed healthy helps his case as MVP, not hurts it as you are implying.

Now with Soriano running like he is, he might be the MVP even on a per game basis. Although his lapses on D may counter some of that.

Similarly, what would our record be against all the right handers if we had Pie instead of Edmonds? Blanco instead of Soto? Or the domino effect of not having a DeRosa?

We have a lot of MVPs on this team.

Harden
Harden has been great, but attributing any of the Cubs non-Harden-start games to Harden is a biiiiig stretch. If anything, his short outings have taxed the bullpen and made the other starts more difficult. That is more than offset by the quality of his own outings every 5 or 6 days, of course.

In fact, Shark has effected more games, appearing in 15 vs 9 for Rich. While Rich has effected the games he was in much more via innings pitched. Meanwhile they have pitched about as well as each other.

Whats more, the fact that Shark replaced an ineffective pitcher makes the net improvement that much bigger, not less important. Harden replaced Gallagher, a big improvement, but not as big as Shark/Howry.

Reed Ballgame - best CF in the MLB

by californiachicagoan on Aug 30, 2008 1:39 PM CDT up reply actions 0 recs

just out of curiosity

let’s take this hypothetical for you

If the Cubs had those same records you indicate with Soriano, Harden, etc but we knew there was some crazy scheduling quirk where the Cubs had played all of those games that Soriano and Harden were in against say the Nationals, Pirates, Padres (the worst teams in the league)

and all the other games that they didn’t play in were against the Brewers, Phillies, Diamondbacks, Cardinals, etc (the best teams in the league)

would you completely ignore the strength of schedule differences in creating the W/L disparities?

I’m not saying this is the case, I’m simply saying to make the decree that W’s and L’s are ALL that matter when analyzing a player’s impact on the team would appear short-sighted and ignoring context. That’s why stats like run differential, strength of schedule, etc are quite useful. They give some context.

If the Cubs were playing .700 ball with Sori and .500 ball without him but we were scoring less runs/game and giving up more runs/game (unlikely event granted), would you really be secure in labeling Soriano the sole difference maker?

by DartmouthCubsFan on Aug 31, 2008 9:40 AM CDT up reply actions 0 recs