The July Juggernaut Effect
As exciting (and necessary) as today's win was, it got me thinking about some of the more established tenets of sabermetrics and how I feel that some of them are conceptually flawed. Now, I'm as big a believer of stat based decision making as anyone else here, but there are few of these ideas that I find faulty, or at least not worth the weight we all put on them. I'll talk about the first in this diary, and I'm curious about what people think.
1. Run Differential as a means of measuring the quality of a team
I understand that over the course of a year, the best teams will score far more runs than they will allow. I also understand that some teams will get a bit lucky and win more games than they should. However, this Royals series got me thinking about a problem I've had with this idea as it has pertained to the A's of the last three or four years.
During this series, we outscored the Royals 18-14. However, we only won 1 of 3 games. That happens sometimes; three games is a small sample size. But anyone who watched those games knows that we played like crap during games 1 and 2 (causing AN to go into a tailspin) and then played abnormally well in game 3. You could argue that this was an ugly series, or at least 2/3 of it was. You could also argue that our 7-0 win today is more indicative of the quality of this team than were the ugly losses, but would that be true?
If we continue to play like this over the course of the year, we will likely end up with a good run differential, and we might even make the playoffs by winning some of these series, but are we then a good team? I mean, a really good team?
This series to me is a microcosm of what the A's have been doing for years. Lose heartbreakers by one or two runs (due largely to an inept offense/bullpen) and then blow out bad teams, often during torrid summer runs. This creates for us a good run differential, but does it make the A's a legitimately good team? Does it make the A's a great team? Again, the numbers suggest yes, but our eyes would tell us a different story.
Could this be why we flamed out in the playoffs every year? Do we artificially inflate our statistics during games like this while losing games that a better team would win? Do we puff up our stature as a team by becoming a July juggernaut and beaing up on the KCs and Tampas of the world?
What do you all think?
0 recs |
3 comments
Comments
Speaking of Juggernaut
by Devyn on May 31, 2006 7:49 PM PDT reply actions 0 recs
I'm guessing that if you looked at this all
Do we excessively lose the close games, or do you just remember such losses more than similar wins? Do we blow out the bad teams more than other playoff-bound teams? I doubt it.
As for our playoff experiences. Those were some hard luck losses. If you think back, you'll find few came down to bullpen, and the one key time it did, we had our best closer ever (Foulke).
Close playoff losses (year by year):
'00
G#3: Balls bounced off plate vs Huddy; too much El Duque
G#5: Heredia digs too big a hole
'01
G#3: Slide; to much Mussina
G#5: Sloppy crap; Meyers throw to first; take 3rd Miggy
'02
G#1: Huddy not right; ugh, Ted Lilly
G#5: Too much Radke; ugh, Billy Koch
'03
G#3: Don't tag the plate; won't keep running
G#4: Can't score off Burkett; Foulke implodes in 8th
G#5: Pedro and Manyy; Lowe vs. Melhuse/Long at the end
Those aren't really bullpen losses in most cases. And in a just world the A's would have one several of those.
Look at it this way: I sometimes play table baseball tournaments in which out of about 20 teams, 4 teams make the playoffs and then play standard series for the title. I won 7 seven straight divisions over a two-year period and lost in the playoffs seven straight times. These were always different teams. I don't think it had much to do with identity since it was a table game. It was just losing 7 coin flips since the playoff teams were so even.
That's sort of what happened to the A's four straight years. It's not much deeper than that, no matter what anyone says.
by RLangford on May 31, 2006 9:24 PM PDT reply actions 0 recs
Record in 1-run games
* only Atlanta (10-13) and the Mets (16-6) have played as many 1-run games; teams with worse records in those games are (not surprising) a bunch of crappy teams (e.g., Pittsburgh, Florida)
2005 - 26-24
* top 10 in number of games; middle of the road in terms of record in those games
2004 - 33-19
* Most games, most wins, second best record (LAD, 32-16)
So in three seasons, one excellent, one bad, one middle of the road. Kind of what you'd expect from a random event. I'm pretty sure there's been extensive BP analysis showing that 1-run games are essentially random. It's one of the reasons some analysts expected the A's to drop last season after an unusually good record in 1-run games the prior year.
by boilerdan on Jun 1, 2006 9:23 AM PDT reply actions 0 recs

by 




















