EXP W/L

Can anyone explain to me how "Exp W/L" under the expanded standings is calculated? I suppose that the run differential has something to do with it. How? Is there more to it than that?

Thanks

I did a google search and found this..........................
---------------------------------------------------------------------------------------------------------------------------------

And to the far right, there's a column titled "Exp. W/L". That is the expected won-loss record for your team, based on Bill James' Pythagorean theorem of baseball. There's a link to a Wikipedia article on it. But essentially it deals with runs scored and runs against, and how that converts to expected wins and losses. If your team is off to either side, it is said to say something about how clutch your team is or isn't.


*ExW-L and ExWP are derived from Bill James' Pythagorean theorem of baseball: Runs scored [squared] / (Runs scored [squared] + runs allowed [squared]). This formula was designed to relate a team's runs scored and runs allowed to its won-lost record.

It's a statistic to tell us we lost because we suck or we won because of pure luck.

Thanks for your help in providing the info Paul. Mr. James is smarter than all of us, so I'll assume it does mean something.

And thanks to Scumby for his concise description of what the formula means to us in reality.

In his original narrative on expected wins the number was supposed to represent the number of wins and losses a team "should" win given the runs scored and runs allowed. The calculation was RS squared / RA squared. If a team won more than the calculation they were lucky. So if a team was over their expected wins mid-season one could expect them to cool off second half as luck evened out. Or if they were below they could be expected to improve the second half as luck evened out. In theory it could be used from one season to the next to predict results. A team playing well above or below their expected wins in one year could expect a reversal of fortune the next as luck was seen to have played a role the previous season.

Several years after the initial formula and as the roid era got in full swing and scoring went up it was determined the formula was not producing win totals that were accurate. If you take the expected wins and losses for every team in a league and total them you would expect the result to be expected wins and expected losses to be equal. But they were not. This led to the realization that as runs increased the formula broke down and the formula was adjusted to use 1.8 as the exponent instead of 2. Logically though as scoring comes back down one would think it would trend back toward the original formula being correct.

My point in all that is that the expected wins is not an absolute and accuracy depends on how well the exponent matches up with the overall scoring level. I would argue that in ATG for example where there are so many great hitters and scoring is taken to a new level even above the steroid era that the formula might need tweaking even beyond the 1.8.

So bottom line don't get hung up too much on expected wins. It is just a concept and a moving target. It is not a hard fact.

Actually, the formula is (RS squared) divided by (RS squared plus RA squared)

luck does play a part

my expected w-l 93-69
actual 82-80
lead the league in run differential and missed the playoffs

I killed lesser teams at home and my division opponents beat me like a red headed step child.

http://onlinegames.strat-o-matic.com/le ... ded/347436

a lesson in free agent pre-season pick ups to fit your division

luck does play a part

my expected w-l 93-69
actual 82-80
lead the league in run differential and missed the playoffs

I killed lesser teams at home and my division opponents beat me like a red headed step child.

http://onlinegames.strat-o-matic.com/le ... ded/347436

a lesson in free agent pre-season pick ups to fit your division

It seems to me a team's record in 1-run games means a lot more than run differential.
Looking at your schedule, there were 6 victories where you scored 15 or more runs (including a 29-3 game) and only 1 loss when you gave up 15 or more runs. the run total for these 7 games is 112-36 or +76
You also won 6 games scoring 12,13 or 14 runs and lost 3 games giving up 12 ,13 or 14. The run total for those 9 games is 100-57 or +43.
So these 16 games you went 12-4 and produced a run differential of +119 (your season total). You were even the other 147 games combined going 70-76.

I don't consider the stat very useful to Strat, at least not like it is for real life MLB.

I feel like my teams win a good number of 1-run games, but if I go with a "small ball" approach, if the opponent gets past my SP, the bullpen is often loaded with guys ready to put gasoline on the existing fire, and turn a 7-4 loss into a 13-4 loss...