Defense Blows Game Postgame

Hmmm. Well, it sounds like there is indeed an effort to measure and include those things in that model. I am curious how in that model managerial decisions, in game strategy of all types are measured and given a numerical value ?

I honestly don’t know and am curious ….For example, the Angels now have a new manager. How would that be accounted for in this model ?

Well, I looked at the FG methodology and it seems unclear in one or two places and very simplistic overall…

I did not see any information about including data related to the manager, strategy etc… In fact, in regards to my question of how this model would account for the fact for instance that the Angels have a new manager, I found only one thing…

This is how they described their methodology : “We have four different projection modes. These determine our projections of each team’s strength..” They then added “All the other factors are held constant.” This would seem to be saying that if a team had to change it’s manager (like the Angels), or a manager was fired, or a team had to move to another place because their stadium was damaged by an earthquake etc, that NONE of these factors would be taken into account.

This seems like a pretty big weakness to me and thus a major limitation of this model.

Further, the fact that they seem to place alot of emphasis on depth is problematic. While we would all generally agree that depth is important and generally helps, it may or may not help any specific team in any situation… For instance, a team with poor depth may not be punished for that if they are generally lucky and don’t lose key starters to injury.

I know there are statistical models that balance these things out, but they are giving credit a priori to teams for depth that they may not even use or benefit from. That is another problem as far as I can see…

Lastly, they say that their Playoff Odds are determined by these three things:

1) current standings
2) remaining schedule
3) each team’s projected performance

What they don’t explain is how # 3 is calculated. Do they mean to say that they soley use the current winning percentage and the remaining schedule to determine the projected performance ? I would guess not but it is not clear…. There had been previous reference to Zips and depth analysis etc…

So, I want to know HOW exactly “each team’s projected performance” is calculated.

You can run 20,000 simulations but if the data you are using is limited or not especially significant then those projections will still be flawed.