Learn how probability, math, and statistics can be used to help baseball, football and basketball teams improve, player and lineup selection as well as in game strategy.

Loading...

From the course by University of Houston System

Math behind Moneyball

31 ratings

From the lesson

Module 4

You will learn how to evaluate baseball fielding, baseball pitchers, and evaluate in game baseball decision-making. The math behind WAR (Wins above Replacement) and Park Factors will also be discussed. Modern developments such as infield shifts and pitch framing will also be discussed.

- Professor Wayne WinstonVisiting Professor

Bauer College of Business

In this video, we're going to learn how to adjust a batter, or, of course,

a pitcher's, statistics based on the part they play, using ESPN's Park Factors which

are not totally accurate, but they're a good first approximation.

So, you probably know Coors Field in Denver, a lot of home runs are hit and

a lot of runs are scored because the air is thin, and

the ball will just rocket right out of it.

As a matter of fact, the park factor for

Coors field works when we're concerned with minute for runs is 1.5,

meaning 50% more runs are scored in Coors Field than an average Ballpark.

At the other end of the spectrum, Safeco Field, this is based on 2014 and

Petco Park in San Diego, Seattle Safeco Field,

roughly 17% less runs per score of the average park.

So, what's the ESPN definition of park factor?

It has some flaws like per runs, you take the total runs

scored per home game by both teams in that park, and

then take the total run score involving that team in their away games, basically.

So, if you're looking at the Colorado Rockies, you'd take the total runs scored

in Coors Field, suppose there's 81 home games home games, divide by 81.

Okay.

You might get five, maybe five runs per game in Coors field, and

then when the Rockies played away,

maybe you get four runs per game and that would get your point factor 1.25.

Well, let's say it's six runs per game at home, and four runs per game on the road.

Then you get a par factor for runs of six over four,

which would be one point five exactly.

Now the problem, what's wrong with this is, you're assuming that

the away games are played in parks of

average park factors, which won't be true because Coors Field being above average,

the rest of them can't be average, that's one [INAUDIBLE].

There are better ways to come up with park factors, but

let's just use what's on ESPN.

I'm sure the errors caused

by the park factors not being perfect are pretty small.

So, let's take, again, Mike Trout 2014,

and let's adjust his statistics based on the park.

Well, the away statistics you wouldn't adjust.

You'd assumed he plays in average parks.

You could adjust every game based on the park factor for that park.

I don't think we want to go that route.

The home stats, here's Angel Stadium's park factors, and ESPN has it for

home runs, hits, doubles, triples, and walks, I guess, walks plus hit by pitcher.

Okay. So,

if you've got a home run factor of 0.5, what does that mean?

You should [INAUDIBLE] for Angel Stadium, it's not 0.5, but that would

mean you would double the number of home runs because he's playing apart,

Mike Trout, that would be twice as difficult to hit home runs in.

So, if I want to adjust Mike Trout's hits, well,

the at bats and plate appearances we're not going to adjust.

So, his adjusted home stats would come from here.

Okay.

By the way, Fan Graphs has the home away splits if you just go to your splits for.

Okay, now so we can adjust the hits.

There are slightly less hits in Angel Stadium, so you would take the hits, and

you would divide by the 0.98.

So, that becomes 84.

It's the doubles you don't adjust at all, but because it's park factor is one,

it stays the same.

So, I would take this divide it by 1,

the triples get bumped way up because there are roughly

half as many triples in Anaheim as an average park,

the home runs, which is critical,

will get bumped up from 19 to probably 24 or so.

Okay.

The walks by hit by pitcher, I'm not including

intentional walks, get bumped up by this.

Okay. Now, since we've got the hits being 84.5,

now the singles would just be the projected hits, minus the doubles,

minus the triples, and minus the homes.

So, I would think you would get at home a couple

less singles because we bumped up everything else by more.

Okay, so now the adjusted total I would add these two rows.

Then when I'm going to evaluate anything involving Mike Trout, or

any other player comparing my track to another player, it'd be best to

look at these at just the totals because they adjust for the perfect play.

So, in other words, I would have them in 39.7 home runs, instead of 36 home runs,

etc.

Okay.

So, that would be the adjusted totals for Mike Trout.

He was really a better player than,

basically, his stats would indicate.

Okay. The fact that he hit 36 home runs

is really like hitting the equivalent of hitting 40 home runs.

So, somebody who hit 36 home runs at an average park would not be considered as

great a power hitter as Mike Trout based on last year.

So, that's how park factors in.

Coursera provides universal access to the world’s best education,
partnering with top universities and organizations to offer courses online.