Page 6 of 6

PostPosted: Thu Apr 27, 2006 5:55 pm
by Bud Light
Just to throw in my $0.02...

There is already data out there for park factors for calculating this stuff. I remember finding it last year and I think Bill James or somebody has it all in one of their books.

Just b/c a guy like Clement has a high hit rate doesn't mean he's a good buy. Once his hit rate gets back to normal he'll probably still be pitching at 4.00+ ERA. Without doing complex calculations you can just compare ERA to WHIP to get a poor man's analysis. If WHIP is 1.00 and ERA is 6.00, you can expect that ERA to come way down. If you want middle class analysis, go to ESPN.com and look at a guys DIPS in their stat section under saber and compare that to current ERA. Then assume the pitcher's ERA will gravitate toward that DIPS. I guess a rich man's analysis includes park factor, K rate, hit rate, line drive rate, etc. As you can see, I'm more of a middle class person. Also, I think there is a better correlation between previous years K rate and ERA more than anything else but I don't have numbers to back it up.

There are some pitchers who are below 30% hit rate for their career and the hypothesis is that it was because they were exceptional fielding pitchers.

You can't extrapolate this theory to the reverse and find buy low hitters if you're using 30% as a yardstick. Batters have their own hit rate over their career which can be above or below 30%. I guess you could find buy low hitters if you compare their career hit rate to their current rate, but that's assuming their K rate and HR rate are the same. However, you could look at Andrew Jones last year who's hit rate was lower than normal last year, yet he had a huge year.

If there's one thing I've learned while trying to find the holy grail of stats, it's that the guys you want to have are the ones with the most talent. Historical data won't tell you future data.

PostPosted: Thu Apr 27, 2006 6:49 pm
by The Loveable Losers
You can't extrapolate this theory to the reverse and find buy low hitters if you're using 30% as a yardstick. Batters have their own hit rate over their career which can be above or below 30%. I guess you could find buy low hitters if you compare their career hit rate to their current rate, but that's assuming their K rate and HR rate are the same. However, you could look at Andrew Jones last year who's hit rate was lower than normal last year, yet he had a huge year.


That's pretty much what I do with hitters. If a guy has sustained a higher hit rate over several years I'll use that for my analysis. The hit rate isn't really affected by changing K or HR rates so you can use it to examine the effect that those things changing have on a player's production. And Jones' big year would be due to the number of balls that left the park (hence making hit rate not nearly as important).

If there's one thing I've learned while trying to find the holy grail of stats, it's that the guys you want to have are the ones with the most talent. Historical data won't tell you future data.


I would definitely agree with you but talent isn't as easy of a thing to measure. Traditional scouting data should definitely be examined though in addition to any statistical information. Either one without the other doesn't give the full picture.

PostPosted: Thu Apr 27, 2006 6:56 pm
by Bud Light
That's pretty much what I do with hitters. If a guy has sustained a higher hit rate over several years I'll use that for my analysis. The hit rate isn't really affected by changing K or HR rates so you can use it to examine the effect that those things changing have on a player's production. And Jones' big year would be due to the number of balls that left the park (hence making hit rate not nearly as important).


I disagree, if you look at data for hit rates there is correlation between more K's and a lower hit rate. The correlation seems to be less for HR's but it still seems to be there. I don't know the reason, but I'd guess that if you're striking out more it is a sign that your strikezone judgement is worse which leads to less quality hits...but it's just a guess. As we know correlation is not causation, so you're technically correct, changing K rate and chaning HR rate don't cause a change in hit rate, but there is a correlation there.

PostPosted: Thu Apr 27, 2006 8:25 pm
by The Loveable Losers
Bud Light wrote:
That's pretty much what I do with hitters. If a guy has sustained a higher hit rate over several years I'll use that for my analysis. The hit rate isn't really affected by changing K or HR rates so you can use it to examine the effect that those things changing have on a player's production. And Jones' big year would be due to the number of balls that left the park (hence making hit rate not nearly as important).


I disagree, if you look at data for hit rates there is correlation between more K's and a lower hit rate. The correlation seems to be less for HR's but it still seems to be there. I don't know the reason, but I'd guess that if you're striking out more it is a sign that your strikezone judgement is worse which leads to less quality hits...but it's just a guess. As we know correlation is not causation, so you're technically correct, changing K rate and chaning HR rate don't cause a change in hit rate, but there is a correlation there.


Interesting...I haven't seen one on some of the players that I've looked at that significantly reduced their strikeout rate. That said I could see it as something that's certainly possible.

Also I tend to take extremely high or extremely low hit rates with a grain of salt. While it's certainly possible that a player can maintain a given hit rate it's also possible they were just lucky. That's the tricky part (especially with young players). How much of a hit rate for a batter is due to skill and how much is due to luck is something that is best determined through traditional scouting techniques as the numbers alone can't tell you the answer to that.

PostPosted: Thu Apr 27, 2006 8:40 pm
by Bud Light
I took a look at last years hot streaks...i defined it as 14 days with a BA greater than .350, and it turns out that the guys who fall into that category have hit percentages within 5-10% of their career averages at the start of their hot streak (for the majority of guys). This is probably getting way off topic for this thread which is about pitchers so i'll stop, but feel free to PM me or start another thread.

PostPosted: Thu Apr 27, 2006 8:49 pm
by HiddenIdentity
Johan Santana may still be a buy-low candidate.

I just made this trade:

Kent/Johan for my Soriano/Izzy.

PostPosted: Thu Apr 27, 2006 8:50 pm
by bceagles04
HiddenIdentity wrote:Johan Santana may still be a buy-low candidate.

I just made this trade:

Kent/Johan for my Soriano/Izzy.


awesome buy low