BtB Power Rankings: 09/24/09

by JinAZ on Sep 24, 2009 10:51 PM EDT in Power Rankings Comment 23 comments

See the rest of the rankings below the jump!

"On Paper" Playoff Leaders (asterisks indicate new leaders):

American League: E=Yankees, C=White Sox, W=Angels*, WC=Rays*
National League: E=Phillies, C=Cardinals, W=Dodgers, WC=Rockies

This Weeks' Movers

"Moving" this week are the power rankings, which have gotten a bit of a facelift this week based on input from Justin Bopp and Sky Kalkman!

There has been no change to the underlying methodology, which was originally present in the first post in this series. The biggest change, aside from the pretty graphics, is that I've renamed eW% to TQI (Team Quality Index), which is the term I'll use for it moving forward. In a sense, I think eW% has always been confusing. While the number is calculated as a "winning percentage," it has never been a number that should be compared to a team's actual winning percentage. Instead, it's a number that represents an estimated winning percentage in a hypothetical league containing all MLB teams--AL and NL alike. AL teams will have a higher TQI than NL teams, on average, because AL teams currently play in a league with a superior level of competition, and therefore we give their component statistics a boost to reflect this difference in league quality. In other words, if we let AL teams play NL teams all season long, they'd end up with higher winning percentages because they'd get to play the inferior NL teams more often).

Below this commentary, you will see tables that break down where the TQI number comes from. First is a table showing expected vs. actual run scored and allowed totals, and how those winning percentages are broken down into estimated winning percentages. If you're interested in seeing why your particular team is ranked differently than you'd expect, this is the place to start. Is it just that the team is deviating from Pythagoras? Or is it that our estimated runs scored or runs allowed numbers are substantially different from what has really happened?

Next are tables breaking down the specific components that feed into our expected run scored and runs allowed numbers. Here you'll find various measures of offense, including measures of clutchiness on offense and differences between ERA and FIP or tRA for pitchers. Feel free to make suggestions of other stats you'd like to see here.

I hope you like the new rankings. Please do let me know what you think.

(next week, this space will include discussion of teams once again--but let me just note that the team surging the most in the rankings right now is my Cincinnati Reds, up two slots and 14-points in TQI!).

Converting Runs to Wins

Team	RS	eRS	RA	eRA	W%	pW%	cW%	LgAdj	TQI
ARI	648	658	705	639	0.431	0.461	0.514	-19	0.487
ATL	693	694	606	642	0.539	0.562	0.536	-19	0.510
BAL	687	670	815	804	0.395	0.418	0.413	19	0.438
BOS	780	786	645	666	0.596	0.589	0.579	19	0.603
CHA	659	654	677	655	0.477	0.487	0.499	19	0.526
CHN	650	654	610	627	0.517	0.529	0.520	-19	0.493
CIN	607	585	675	689	0.467	0.451	0.425	-19	0.398
CLE	725	737	809	796	0.404	0.447	0.463	19	0.487
COL	695	708	618	569	0.566	0.554	0.599	-19	0.572
DET	692	687	685	698	0.536	0.505	0.492	19	0.518
FLA	746	749	741	743	0.536	0.503	0.504	-19	0.480
HOU	616	625	717	722	0.467	0.430	0.433	-19	0.407
KC	645	621	768	723	0.414	0.418	0.430	19	0.456
LAA	836	810	726	773	0.592	0.568	0.523	19	0.546
LAN	759	763	578	607	0.599	0.625	0.607	-19	0.581
MIL	722	719	751	759	0.493	0.481	0.474	-19	0.450
MIN	744	747	713	755	0.520	0.520	0.495	19	0.519
NYA	867	909	717	713	0.634	0.592	0.618	19	0.640
NYN	631	657	728	742	0.425	0.433	0.442	-19	0.417
OAK	727	686	724	670	0.474	0.501	0.511	19	0.537
PHI	747	748	635	666	0.583	0.576	0.555	-19	0.530
PIT	604	591	741	728	0.373	0.405	0.404	-19	0.379
SD	658	657	802	774	0.458	0.406	0.422	-19	0.397
SEA	619	622	676	697	0.520	0.459	0.447	19	0.473
SF	612	572	577	621	0.539	0.526	0.463	-19	0.434
STL	701	705	604	634	0.582	0.568	0.550	-19	0.523
TB	761	788	702	663	0.513	0.538	0.582	19	0.606
TEX	707	684	661	656	0.550	0.531	0.519	19	0.545
TOR	720	734	708	703	0.454	0.508	0.521	19	0.546
WAS	659	687	821	820	0.344	0.396	0.416	-19	0.392

Table Legend

RS = Actual Runs Scored, after a park adjustment
eRS = Estimated Runs Scored, after park adjustment (see "Offense" table below)
RA = Actual Runs Allowed, after a park adjustment
eRA = Estimated Runs Allowed, after park adjustments (see "Defense" table below)
W% = Actual Winning Percentage
pW% = PythagenPat Winning Percentage, based on actual runs scored and run allowed totals
cW% = Component Winning Percentage (previously eW%lg), using estimated runs scored and estimated runs allowed totals
LgAdj = League adjustment, based on differences in league quality (justification here and here).
TQI = Team Quality Index, a hypothetical winning % based on component estimates of runs scored and runs allowed after the league adjustment.

Team Offenses

Team	RS	eRS	wOBA	wRC	EqBRR	Clutch
NYA	867	909	0.364	915	-6	-0.78
LAA	836	810	0.349	811	-1	1.99
TB	761	788	0.347	785	3	-3.2
BOS	780	786	0.345	787	0	-2.52
MIN	744	747	0.338	742	5	-2.29
PHI	747	748	0.338	750	-2	3.19
LAN	759	763	0.338	767	-4	-3.62
FLA	746	749	0.337	749	0	1.04
CLE	725	737	0.336	740	-3	-5.21
TOR	720	734	0.334	725	9	-7.66
MIL	722	719	0.333	728	-10	1.53
STL	701	705	0.332	702	3	1.27
COL	695	708	0.331	701	6	-3.96
TEX	707	684	0.330	686	-2	-1.86
DET	692	687	0.330	689	-2	1.83
WAS	659	687	0.328	696	-9	-4.49
ATL	693	694	0.327	708	-14	-1.68
OAK	727	686	0.327	674	12	-3.34
BAL	687	670	0.324	691	-21	-2.61
SD	658	657	0.322	665	-8	2.71
NYN	631	657	0.322	653	3	-1.1
CHA	659	654	0.322	663	-8	-2.48
CHN	650	654	0.320	667	-13	-5.25
ARI	648	658	0.319	664	-6	-3.99
HOU	616	625	0.319	624	1	3.44
SEA	619	622	0.317	623	-2	2.32
KC	645	621	0.317	632	-11	-1.65
PIT	604	591	0.313	600	-9	-2.79
CIN	607	585	0.308	593	-8	-2.11
SF	612	572	0.308	561	11	3.48

Table Legend

RS = Actual Runs Scored
eRS = Estimated Runs Scored: wRC + EqBRR
wOBA = The Book's statistic, but park adjusted, and using data from both wRC and EqBRR
wRC = From FanGraphs, with baserunning removed, after park adjustments
EqBRR = Dan Fox's fielding composite fielding statistics from Baseball Prospectus
Clutch = "Clutchiness" measure from fangraphs; difference between actual WPA and expected WPA based on component statistics.

Team Defenses

Team	RA	eRA	ERA	tERA	tRuns	BABIP	Field	Catch
COL	618	569	4.25	3.56	574	0.307	7	-2
LAN	578	607	3.40	3.78	629	0.280	25	-3
CHA	677	655	4.21	3.82	623	0.303	-27	-6
ATL	606	642	3.62	3.85	629	0.308	-17	3
CHN	610	627	3.85	3.88	629	0.293	-4	7
ARI	705	639	4.41	3.95	648	0.308	20	-11
BOS	645	666	4.24	3.96	636	0.319	-22	-8
STL	604	634	3.62	3.97	646	0.297	4	9
SF	577	621	3.62	4.03	654	0.291	39	-6
KC	768	723	4.72	4.10	658	0.314	-50	-16
OAK	724	670	4.28	4.12	671	0.308	-5	6
TOR	708	703	4.43	4.24	692	0.313	-18	7
PHI	635	666	4.09	4.30	698	0.300	27	5
FLA	741	743	4.36	4.36	712	0.311	-27	-4
NYA	717	713	4.35	4.36	715	0.299	4	-2
HOU	717	722	4.46	4.36	703	0.315	-22	2
TEX	661	656	4.33	4.37	701	0.291	40	4
TB	702	663	4.30	4.40	705	0.297	40	2
NYN	728	742	4.53	4.42	712	0.305	-37	7
MIN	713	755	4.52	4.47	724	0.308	-31	-1
CIN	675	689	4.22	4.54	744	0.289	43	11
DET	685	698	4.33	4.62	739	0.298	27	14
PIT	741	728	4.67	4.63	733	0.305	11	-7
SD	802	774	4.46	4.67	766	0.303	-10	2
LAA	726	773	4.50	4.72	765	0.306	-1	-7
SEA	676	697	3.91	4.72	771	0.279	67	7
CLE	809	796	5.08	4.75	761	0.313	-33	-2
MIL	751	759	4.76	4.77	768	0.301	12	-3
BAL	815	804	5.15	4.77	767	0.316	-40	3
WAS	821	820	5.17	5.00	788	0.306	-23	-9

Table Legend

RA = Actual Runs Allowed, after park adjustment
eRA = Estimated Runs Allowed: tRuns - Field - Catch
ERA = Straight-up Earned Run Average
tERA = Estimated Earned Run Average, a home brew version of Graham McAree's statistic)
tRuns = Pitching Runs Allowed, based on tERA
BABIP = Straight-up Batting Average on Balls In Play
Field = An average of bUZR from FanGraphs and THT's team fielding statistic, converted to runs.
Catch = Catcher Fielding Runs, based on SB's, CS's, WP's, PB's, and E's, described here

Methods underlying these rankings were described in more detail in the first post in this series.

Poll

Do you like the new look of the Power Rankings?

Yes, nice improvement!

75 votes

Meh ::shrug::

24 votes

Bring back to the old.

11 votes

110 votes | Poll has closed

2 recs | Comment 23 comments | Share

Story-email Email Printer Print

Comments

HOLY CRAP!

This is fantastic!

"What we do in life, echoes in eternity!"

by Justin Bopp on Sep 24, 2009 11:20 PM EDT reply actions 0 recs

The thing I like most about it...

…is that by dividing up the tables, I actually have room to post more data each week. :) So, now I can get wRC, EqBRR, Fielding, and the Catching stats all with their own columns. Before, I was running out of space.
-j

I write at:
Beyond the Boxscore
Red Reporter
Basement-Dwellers.com
Twitter: @jinazreds

by JinAZ on Sep 24, 2009 11:32 PM EDT up reply actions 0 recs

You can post more data and yet it's easier to take in.

Start really basic with the rankings. Then break it down into actual vs. expected. Then break expected into all the smaller pieces.

Beyond the Boxscore Not a member? Sign up.

by Sky Kalkman on Sep 25, 2009 8:25 AM EDT up reply actions 0 recs

Nitpick: clutch doesn't need so many decimal places.

These are awesome, really awesome.

Beyond the Boxscore Not a member? Sign up.

by Sky Kalkman on Sep 25, 2009 8:48 AM EDT up reply actions 0 recs

Yeah, fair enough

I’m reporting it as it is on FanGraphs, but I’ll drop it down.

Is Clutchiness reported as runs or wins? I think wins, as it’s compared to WPA. If wins, I’ll leave a tenth. If runs, I’ll round to the whole number.
-j

I write at:
Beyond the Boxscore
Red Reporter
Basement-Dwellers.com
Twitter: @jinazreds

by JinAZ on Sep 25, 2009 10:38 AM EDT up reply actions 0 recs

If it's wins, can you convert to runs?

Keeps everything on the same scale.

Beyond the Boxscore Not a member? Sign up.

by Sky Kalkman on Sep 25, 2009 11:18 AM EDT up reply actions 0 recs

It is in Wins

And sure, why not convert to runs? I’ll even use r/g instead of 10 runs/win to make it more accurate. :P
-j

I write at:
Beyond the Boxscore
Red Reporter
Basement-Dwellers.com
Twitter: @jinazreds

by JinAZ on Sep 25, 2009 12:20 PM EDT up reply actions 0 recs

I still get confused with NYN and NYA

Where did NYN and NYA originate? I always wondered?

I know “N” stand for National League and “A” is for the American League.

NYN = Mets

NYA = Yankees

by nelsonc on Sep 25, 2009 12:04 AM EDT reply actions 0 recs

Agreed.

Why change the universally accepted abbreviations?

NYY
NYM

No questions.

"What we do in life, echoes in eternity!"

by Justin Bopp on Sep 25, 2009 12:18 AM EDT up reply actions 0 recs

I think it comes from the way the Baseball Database names the teams.

There, they show up as NYA, KCA for KC Royals, SFN etc. for any team with only two letters in the location abbreviation.

---
Beyond The Boxscore
twitter

by Jack Moore on Sep 25, 2009 12:55 AM EDT up reply actions 0 recs

Right, these abbreviations carry over from my data sources

E.g. here:
http://www.hardballtimes.com/main/teams/

I’m accustomed to these abbreviations and I’m surprised that folks are struggling with them. But I can fix it with some nested vlookups, so I can make that adjustment for next week. I’ll convert to B-Ref’s abbreviations, ok?
http://www.baseball-reference.com/
NYY, NYM, CHW, CHC, SFG, etc.
-j

I write at:
Beyond the Boxscore
Red Reporter
Basement-Dwellers.com
Twitter: @jinazreds

by JinAZ on Sep 25, 2009 8:41 AM EDT up reply actions 0 recs

I wouldn't go crazy

The N/A designations are fairly commonplace, IMO.

by Eric Simon on Sep 25, 2009 9:24 AM EDT up reply actions 0 recs

I think so too

But there’s apparently a sentiment to change them here. I don’t think anyone will complain about moving to B-Ref’s abbreviations.

I write at:
Beyond the Boxscore
Red Reporter
Basement-Dwellers.com
Twitter: @jinazreds

by JinAZ on Sep 25, 2009 10:39 AM EDT up reply actions 0 recs

I second this

I all ready screwed the two up when I was doing a thing about divisional strength. No way I would screw up NYY or CWS

by lookatthosetwins on Sep 25, 2009 12:45 AM EDT reply actions 0 recs

College World Series?

I don’t have a good answer for you, except you’ll see the N/A abbreviations in a number of places when you go to gather data.

Beyond the Boxscore Not a member? Sign up.

by Sky Kalkman on Sep 25, 2009 8:26 AM EDT up reply actions 0 recs

"This"

means Justin Bopp’s comment about NYY, I apparantly can’t figure that gosh darned reply button out

by lookatthosetwins on Sep 25, 2009 12:49 AM EDT reply actions 0 recs

This?

"What we do in life, echoes in eternity!"

by Justin Bopp on Sep 25, 2009 12:50 AM EDT up reply actions 0 recs

Forgot to mention that I rec'd this article, and you should too. It's time we support good articles around here!

+1!!

"What we do in life, echoes in eternity!"

by Justin Bopp on Sep 25, 2009 12:52 AM EDT reply actions 0 recs

Hey

Digging the new graphics. The only suggestion I’d make is that rather than having a double-sided arrow <→ for no movement, perhaps just a hyphen might be better. If you’re not necessarily looking for any particular team but rather scanning that column for big movers, hyphens might make the column appear a little cleaner.

Other than that though, this is (as always) great stuff.

The Rockies need some oldschool purple/white striped high socks. The team’s problem is it’s lack of swagger. I feel strongly that these socks will provide the swagger necessary to tap the potential that are the Rockies.

by Resolution on Sep 25, 2009 1:08 PM EDT reply actions 1 recs

Thanks

Seems like a good suggestion. I was just so excited about the arrows that I sort of went crazy. :)
-j

I write at:
Beyond the Boxscore
Red Reporter
Basement-Dwellers.com
Twitter: @jinazreds

by JinAZ on Sep 25, 2009 8:59 PM EDT up reply actions 0 recs

+1

"What we do in life, echoes in eternity!"

by Justin Bopp on Sep 26, 2009 12:50 AM EDT up reply actions 0 recs

I have been aware of this series all season but not keenly intimate with the details

Are all ranking based on the team’s statistical output for the whole season? The point: St. Louis is a better team than you’re showing since Matt Holliday has been there for only 1/3 of the season.

"I have no special talents. I am only passionately curious." - Albert Einstein

by Andrew T. Fisher on Sep 30, 2009 11:15 AM EDT reply actions 0 recs

yes and yes

It’s strictly based on team stats to date. That probably does mean that St. Louis is a better team than these measures indicate, given Holliday and everyone else that is now on the team who weren’t playing earlier in the season.

vivaelpujols had postseason odds estimates earlier this year that used these data, plus preseason pecota, plus some manual adjustments for acquisitions. I’m not doing that, however.
-j

I write at:
Beyond the Boxscore
Red Reporter
Basement-Dwellers.com
Twitter: @jinazreds

by JinAZ on Oct 1, 2009 12:45 PM EDT up reply actions 0 recs