For all you analyst geeks such as myself. You always hear about teams record in 1 Run games. What is everyones opinion on it, I for one think it really means nothing.

a) If you have a Winning record say 15-3 (like my current team) are they good or just a hair away from being average and have been lucky.

b) If you have a 3-15 record are you that bad or are you just a few breaks away from a few more Wins

This is different than extra innings games that may indicate you have a strong / deep staff to keep you in games until you Win it but if you Win 8-7 in a 9 inning game or 3-2....does it really matter?

I look at 1-run games as just a measure of luck, in my experience, subtracting 1-run games from a teams record the W% is very close to the pythagorean W% for the same team...

In real life, I'd say it's a measure of an offense's ability to step up and scrape out that extra run, or a pitching staff's ability to bear down and protect that lead at all costs.

Hard to argue for intangibles like that in the SIM though.

mostly luck and small sample size but not entirely. the weaker your bullpen the more likely to have a losing record in 1-run games. speed an advantage in 1-run games. I make it a priority to have enough Setup innings and I like to draft a pinch runner.

Quote: Originally Posted By rbow923 on 1/25/2010
mostly luck and small sample size but not entirely. the weaker your bullpen the more likely to have a losing record in 1-run games. speed an advantage in 1-run games. I make it a priority to have enough Setup innings and I like to draft a pinch runner.

My experiences are the same. Speed and a good bull-pen are keys to 1-run victories.

Bill James has many times said the W-L record in 1-run games is mostly luck.

A team's W-L record in 1 run games has a very small correlation to overall W-L record. But a team's record in 3+ run games has a very strong correlation to overall record.

Here's one division from one of my leagues...

West	W-L	PCT	GB	Division	1-Run Games	Extra Inning	RS	RA	Exp. Pct.	Owner	LIVE
Lucky to win 60	88-74	0.543	-	25-17	19-26	5-8	871	723	0.588	jetscanes
Unbalanced 60	85-77	0.525	3	22-20	23-28	3-10	775	758	0.511	ballantine
Highly Unbalanced	76-86	0.469	12	18-24	21-25	8-1	790	852	0.464	mpitt76
Tainted Disaster	67-95	0.414	21	19-23	27-28	8-4	511	639	0.404	The Taint

Yes, there are certain things you can try to do to improve your chances at winning close games (speed, defense, bullpen), but I'd rather have a good record in blowout games and take my chances at getting lucky at close games, then vice versa.

Agreed, mostly luck. If you have a bullpen problem that should be evident without worrying about 1-run W/L.

Quote: Originally Posted By schwarze on 1/25/2010
Bill James has many times said the W-L record in 1-run games is mostly luck.

A team's W-L record in 1 run games has a very small correlation to overall W-L record. But a team's record in 3+ run games has a very strong correlation to overall record.

WIS is not real life baseball. Luck is more of a factor is 1-run games than in three-run games. However, I believe most experienced players can agree that certain strategies or tactics are condusive to winning 1-run games.

a dominant team should have a winning record in 1-run games too. if they score a lot more than they give up they are statistically more likely to be +1 than -1.

Just to quantify... I took the data from the six leagues from the WIS Championship round 2 (144 teams).

I seperated the W-L% in 1-run games vs. the W-L% in all other games for each of the 144 teams. I then ran a regression analysis to test the correlation.

For those who are not math geeks, correlation is the statistical measurement of the relationship between two variables. It can range from -1 to +1. Zero means no correlation.

Anyway, the correlation coefficient between 1-run W-L% to Overall W-L% was 0.55 for these 144 teams - which does show positive correlation. The correlation between 2+ run W-L% to Overall W-L% was 0.95... almost perfect correlation.

The problem with comparing 1-run or 2+ run W-L% to overall W-L% is that the latter stat includes the former stat. So the best way to measure this is to compare the two measure against each other.

The correlation between 1-run W-L% and 2+ run W-L% is 0.29, a slightly positive correlation (but alot closer to zero than to one).

I stand by my earlier comments, 1-run W-L% is mostly due to luck.

1 Run Games is a bit of a Problem in this Game. And I'll now tell you why.

I recently was playing a very important game against a fellow Wild Card Contender in my League. In this game, I had started my #3 Starter with a fresh Juan Marical(#1 Starter) due to start the next game. However in the 6th ing the game became tied, and an ideal situation was created to bring in Juan. However this wasn't possible and I lost the game by 1 run.

Zubinsum Getting Posterized By Schwarze...