Breaking Down

[bhogg]

I agree --- this kind of stuff is incredible.

 

I also agree that Hendry and his cronies have no [expletive] clue.

[Rob]

I'll clarify a bit, just for you.

 

The process isn't hard. Imagine for a moment you know how many ABs a player has, and their batting average... but you want to know how many hits they had. You simply multiply them together and voila! Well, we're doing the same thing... multiplying the expected batting average on balls in play (which is LD% + .120) by the number of balls in play. That gives you the expected number of hits in that situation.

 

That flipped the light switch to on. Thank you for bringing me out of the dark. Now I feel even more dumb, lol. Nice work, btw, as everyone has said.

 

Don't feel stupid; I did a terrible job explaining what I was doing. Glad to hear you get it now though.

[Sosa21MVP]

Thanks Rob. Solid thread. A heck of a lot more entertaining and informative than reading the latest "so and so sucks" thread.

[Z38Cubs]

OK, so someone called the Boers and Bernstein show today to talk about the expected lines of the Cubs hitters and the bad luck that a couple of them have had. Bernstein seemed intrigued, but Boers chalked it up to more excuse making by the Cubs and their fans. He didn't seem to understand the concept. Anyway, I was wondering if it was someone on this board that made that call.

[Rob]

OK, so someone called the Boers and Bernstein show today to talk about the expected lines of the Cubs hitters and the bad luck that a couple of them have had. Bernstein seemed intrigued, but Boers chalked it up to more excuse making by the Cubs and their fans. He didn't seem to understand the concept. Anyway, I was wondering if it was someone on this board that made that call.

 

Ok, that would be pretty sweet.

[chuckywang]

OK, so someone called the Boers and Bernstein show today to talk about the expected lines of the Cubs hitters and the bad luck that a couple of them have had. Bernstein seemed intrigued, but Boers chalked it up to more excuse making by the Cubs and their fans. He didn't seem to understand the concept. Anyway, I was wondering if it was someone on this board that made that call.

 

To be fair, I doubt most of the posters in this thread know where BABIP = LD + .120 comes from.

[Rob]

OK, so someone called the Boers and Bernstein show today to talk about the expected lines of the Cubs hitters and the bad luck that a couple of them have had. Bernstein seemed intrigued, but Boers chalked it up to more excuse making by the Cubs and their fans. He didn't seem to understand the concept. Anyway, I was wondering if it was someone on this board that made that call.

 

To be fair, I doubt most of the posters in this thread know where BABIP = LD + .120 comes from.

 

To be honest, there's really no reason to. It's used simply because it's a good approximation. If you need more than an approximation, there are better formulas to use that incorporate other factors into the mix.

[madvillian]

I really don't want anyone to take this the wrong way, as this might be like my 5th post here and I'm not a Cubs fan...

 

But -- using LD% + .120 predictively was found to be worse than using actual BABIP by the Hard Ball Times earlier this year.

 

If you believe their research (and with their rep and my rudimentary statistical knowledge, I am) then the premise of this thread is invalid.

 

Thus, there are no conclusions to be drawn from calcuating expected BABIP using this method, and we should expect future BABIP for these players to closer to their actual BABIP rather than the calculated value of LD% + .120.

 

This thread of course could be done over using their best model -- xBABIP.

 

Here is the article:

http://www.hardballtimes.com/main/fantasy/article/whats-the-best-babip-estimator/

 

[edited for clarity]

[Rob]

I really don't want anyone to take this the wrong way, as this might be like my 5th post here and I'm not a Cubs fan...

 

But -- using LD% + .120 predictively was found to be worse than using actual BABIP by the Hard Ball Times earlier this year.

 

If you believe their research (and with their rep and my rudimentary statistical knowledge, I am) then the premise of this thread is invalid.

 

Thus, there are no conclusions to be drawn from calcuating expected BABIP using this method, and we should expect future BABIP for these players to closer to their actual BABIP rather than the calculated value of LD% + .120.

 

This thread of course could be done over using their best model -- xBABIP.

 

Here is the article:

http://www.hardballtimes.com/main/fantasy/article/whats-the-best-babip-estimator/

 

[edited for clarity]

 

You're misinterpreting the purpose of my work slightly. I wasn't attempting to predict values for next season... of course actual BABIP would have a higher correlation for subsequent seasons when taken across such a broad sample size. I'm merely attempting to pinpoint anomalies within this season... a task which this is well suited for.

 

In addition, I am quite lazy. Do you realize it would take me longer simply to calculate xbabip for one batter than it did for me to put all of this together with a simple LD%+.120?

[madvillian]

Rob: Actually your work (well intentioned if flawed) has inspired me to do some of my own. I'm currently working on calculating the expected BABIP for a few teams (including the Cubs) with some database functions using the "quick expected BABIP" developed by the Hard ball times. Not quite as good as xBABIP but still very good.

 

It's proving a little frustrating as I'm trying to reverse engineer a spreadsheet calculator they put together and move it into a database where I can quickly calculate it for entire teams at a time.

 

We'll see how motivated I am over the next few hours, but hopefully I'll have the Cubs by the end of the night.

[Rob]

Rob: Actually your work (well intentioned if flawed) has inspired me to do some of my own. I'm currently working on calculating the expected BABIP for a few teams (including the Cubs) with some database functions using the "quick expected BABIP" developed by the Hard ball times. Not quite as good as xBABIP but still very good.

 

It's proving a little frustrating as I'm trying to reverse engineer a spreadsheet calculator they put together and move it into a database where I can quickly calculate it for entire teams at a time.

 

We'll see how motivated I am over the next few hours, but hopefully I'll have the Cubs by the end of the night.

 

Bah. Flawed is a misrepresentation. Could it be done better? Absolutely. But it still works.

 

See... what you're doing is where I suffer. I can't figure out how to import data directly into the spreadsheets I already have set up... forcing me to type in each and every line by hand each time I want to calculate anything. Best of luck to you though... there's no such thing as too many people crunching numbers. I can't wait to see what you come up with.

[madvillian]

So, I have the formula into my database, just trying to do some data entry on the players and teams I'm interested in looking at. Unfortunately there isn't a good database format spreadsheet or text file to import for the qxBABIP formula.

 

I did run one player quickly though: in one of my favorite baseball convergences, intuition is backed by statistical analyis: qxBABIP has Fukodome at .334 -- 15 points higher than his .319. He's been unlucky. Dude hits a ton of line drives.

 

 

 

I'll try and get you guys most frequent batter totals from 2009 here in an hour or so.

[madvillian]

It's a real pain in the ass to get this in a database usable format. I setup a quick entry form and it still takes about 2-3 minutes per player of dedicated entry using Fangraphs.

 

I did probably the 4 players of most interest to most Cubs Fans: Fukodome, Soto, Soriano and Lee. (Ramirez just not enough PA this year.)

 

Fukodome's expected is .334 and his actual is .319.

 

Lee's expected is .304 and his actual is .322

 

Sori's expected is .313 and his actual is .283

 

Soto's expected is .316 and his actual is .253

 

Soto has just not had any luck. I'm not sure what his arb status is, but ironically, his "off" year this year might perhaps help the Cubs long term.

 

Again, I calcuated these values using "quick expected BABIP" w/ the park adustment's from the Hard Ball Times. Although not as good as xBABIP it is still very good as a predictive model and relies on stats available to the everyday Internet savy fan.

 

If you want more info please see the Hard Ball Times articles on xBABIP.

 

If you want more info on any of the database work I've done PM me and I can send you over the basic structure.

 

I'll do the rest of the top batters for the Cubs soon.

 

If it's any consolation (and it often is) the "Savior", Gordon Beckham's qxBABIP is .296 and his actual is .334. The hype machine in him is no doubt based in reality, but it might nonetheless be a bit premature.

[Flames24Rulz]

OK, so someone called the Boers and Bernstein show today to talk about the expected lines of the Cubs hitters and the bad luck that a couple of them have had. Bernstein seemed intrigued, but Boers chalked it up to more excuse making by the Cubs and their fans. He didn't seem to understand the concept. Anyway, I was wondering if it was someone on this board that made that call.

 

Ok, that would be pretty sweet.

 

I have to be honest, I sort of was thinking about calling up today when I heard that guy call as well.

 

Anyway this is a good thread. It's ridiculous to see how many of Kosuke's outs in play are line drives. Same with Geo.