League Updates Uncategorized

How Good (or Bad) is Your Team, Really?

It’s getting awfully late for the Wolverines and the Balk.  It’s getting harder and harder to believe our teams are just on the verge of snapping out of it and winning like they’re supposed to. It’s also getting harder to believe the Cheese or other top teams are just about to see the balloons pop and come gliding back to earth — but this latter fact doesn’t  concern those of us at the bottom of the league. What happens atop the standings is so far away it might as well be happening on another planet. A 10-game losing streak would not put the Cheese on a collision course with us. It would just mean subtle change in the alignment of celestial bodies we can barely espy with the naked eye — like the slow changes that transform Venus from an evening to a morning star, without anyone taking note — and without ever bringing the planet close enough to be any more than a point in the dark night sky.

Where was I? Oh, yes, it’s getting later for us at the bottom to hold out hope of contending. But what about hope of finishing above .500?  80% of the league is above .500 today. Is there room for the Wolverines and/or the Balk above sea level?

Bill James knows.  Or at least, he did a study about it.  He says there are two factors at play in a team’s record on any given day in the season.  The first is the weight of each win and loss as a fixed result that cannot be changed.

“If the actual quality of the team was not a factor in their won-lost record through 40 games—that is, if the 22-18 teams were in reality no better than the 18-22 teams; they were all .500 teams, but some had been a little bit luckier than others—then

              24% of 16-24 teams would go on to finish over .500,

29% of 17-23 teams,

36% of 18-22 teams,

43% of 19-21 teams,

50% of 20-20 teams,

57% of 21-19 teams,

64% of 22-18 teams, and

71% of 23-17 teams.

The second factor is the evidence each win (or loss) provides that your team is better (or worse) than .500.  A small variance from .500 doesn’t tell you much. As Bill James put it: ”

If your team is 21-19, that indicates almost nothing about the quality of your team.   Your chances of finishing over .500 are 58%–57% because you are two games over .500 so far, and 1% because you are actually good.

For the 1063 teams James studied,  roughly the percentage of record that seemed be due to the teams’ essential quality at 40 games:

16 – 24  — 5 %

17 – 23 —  3 %

18 – 22 —  4 %

19 – 21 —   1%

20 – 20 —  1%

21 – 19  —   1%

22 – 18 —   6%

23 – 17 —   8 %

I don’t know why the data doesn’t describe smoother or more symmetrical curve. But it does say our confidence level that the Cheese are all that much better than the Wolverines is surprisingly low.   Not nonexistent — just not overwhelming, not massive, not as big as it looks from down here.  Or to make the same point prospectively: my shock at the way the Wolverines’ season has turned out so far is not necessarily evidence that I am staggeringly stupid.  You will need more evidence to convict me of staggering stupidity. I will take the fifth on how plentiful this additional evidence is, although I have a disemboweled-but-still-mildewy van on the driveway and two hall lights that no longer turn on at the same time — just to name a couple of unrelated possibilities close at hand.

Anyway… the vast gulf between the Cheese and Wolverines in the standings is real (and generally growing). Those are facts on the ground I can’t do much about.  But it’s still possible that a healthy Wolverine team could beat a healthy Cheese team if we were to start over today.  We aren’t going to do that, and the Cheese will have every right to crow about their trophy (if they win it) as loud and obnoxiously as some… a few… ok, as one of our past champions has been known to do.

 

I don’t know if you’ve noticed yet — it’s sitting there in the intro to this  piece, if you know what to look for — but James’ study confirms the brilliance of the Balk and Wolverine management: we are the only two who have a way of comparing ourselves favorably with MLB  GM’s.  If we manage our teams to a finish over .500, we will have demonstrated the greater-than-50% probability that we can out-manage the average MLB organization.  All we have to do is get over .500 and we’ll have documented that we did better than a defined percentage of the managerial population.  At 18 – 23, only 30% of MLB organizations were able to finish above .500.  If the W’s do it, I’ll be at least a 70th percentile top General Manager.  At DC’s record,   Rob could demonstrate the likelihood that he is in the top 2% of GM’s by sneaking his team across the finish line above .500.

EFL Standings for 2017
EFL
TEAM WINS LOSSES PCT. GB RS RA
Cottage Cheese 30 14 .692 251.5 165.3
Kaline Drive 28 17 .626 2.8 214.1 165.7
Haviland Dragons 28 17 .614 3.3 260.8 207.9
Pittsburgh Alleghenys 25 16 .622 3.5 220.7 169.9
Flint Hill Tornadoes 25 17 .606 4.3
Peshastin Pears 26 20 .556 5.9 213.5 198.3
Portland Rosebuds 26 21 .554 6 256.2 219.6
Canberra Kangaroos 24 19 .547 6.4 222.2 195.5
Old Detroit Wolverines 18 24 .425 11.6 166.1 194.5
D.C. Balk 15 28 .359 14.5 204.5 274.2
.
Cottage: DNP: 1- 0.  Chance of finishing over .500: >95% — Bill James’ charts for 44 games-played only go up to 27-17.  Be careful, Dave, you don’t want to be in the bottom 5th perentile!
.
Kaline: W, 4 – 1.  (.200, .304, .350; 7 ip, 1 er).  Go Michael Fulmer!  The Drive’s chances of finishing over .500:  90%.
.
Haviland: W, 4 – 3. (.250, 333, .792;  7 ip, 3 er)  (Finish above .500:  90%  )
.
Pittsburgh: “L”, 7 – 4.  (.379, .438, .448; 6 ip, 2 er) . (Finish above .500:  92%)
.
Flint Hill:  W , ?? – ??.  (.235, .278, .294, no pitching). RS and RA in the standings are probably wrong. I accidentally loaded the entire season of Tornado stats into the database.  I’ve reloaded the proper stats, which reflect in the standings. But the database doesn’t refresh the games behind when you reload for the secoeend time on a day, and I’m not sure whether it reloads RS and RA.  This will all sort out tomorrow. Although I won’t have a daily score for the Tornadoes then, either.  (Finish above .500:  86%  )
.
Peshastin:  “W”, 5 – 7. (.300, .333, .533, 1 ip, 0 er) (Finish above .500:  79%  )
Portland: W 1, L 1; 6 -4. (.250, .250, .750). (Finish above .500: 73% )
.
Canberra: W 1, L (-1);  (-2) – (-2). (.250, .276, .393;  2.3 ip, 0 er). (Finish above .500:  76%)
.
Old Detroit: L,  4- 6. (.273, .385, .364;  2 ip, 0 er). (Finish above .500:   28% )
.
DC:  DNP, 0 – 0. (.211, .238, .526;  no pitching). (Finish above .500:  <5% — James’ chart only goes down to 16- 27.)

3 Comments

  • Maybe it is just a typo but your summary has the Dragons allowing 3 ERs yesterday. I can only find 1. Am I missing something?

    • My mistake… a typo. However your team in EFL competition was credited with 3 runs allowed possibly rounded of from some fraction.

      • thanks for the quick response. I figured something like that had happened. Thanks for your hard work on our behalf!