The W
Views: 97897296
Main | FAQ | Search: Y! / G | Calendar | Color chart | Log in for more!
2.8.14 0153
The W - Baseball - Baseball Statistics
This thread has 49 referrals leading to it
Register and log in to post!
Pages: 1 2 Next(1489 newer) Next thread | Previous thread
User
Post (24 total)
TheBucsFan
TheChiefsFan








Since: 2.1.02

Since last post: 3 days
Last activity: 1 day
#1 Posted on | Instant Rating: 4.42
What statistic for batters do you think is the most telling sign of a hitters ability or value to his team? While I realise ideally you would have a player with big numbers in several categories, ignore that possiblity for now. Do you want a guy who drives in runs, or do you want someone who can be counted on to put the bat on the ball? Speed or power? Some possibilities:

1)Runs produced (Runs scored+RBI-HR): The problem with this stat is it really relies more on the ability of the hitters around the player than the player himself. A player in a lineup of poor hitters won't have the opportunity to drive in runs or be driven in himself.

2)Batting Average (Hits/At Bats): Simple and telling. How good the guy is at hitting the ball where fielders can't get to it. But some guys excel at getting on base in other ways, and this stat varies with guys who get more or less at bats, so...

3)On Base Percentage ((Hits+Walks)/(At Bats+Walks)): How good is the guy at getting on base? Getting on base of course is the first step to scoring runs. Definitely important. But what about guys who don't get on base as much, but when they do, they advance far on the base paths?

4)Slugging Percentage (Total Bases/At bats): The number of total bases a batter gets per time at bat. Still not revealing as it may seem, though, because a player who singles every time up would have the same as a player with the same # of AB who hits .250, with all hits being a HR. So we have...

5)Extra Base Hits (2B+3B+HR): Not much to say, just that this is the ultimate source of a player's power hitting ability.

6)Strikeout rate (K/AB): How good is the player at putting the ball in play? This is my personal favorite if I'm putting together a team. If players are seemingly identical, I go with who strikes out less often, because if you can't put the ball in play, you are useless. Even this one has its faults though, because a player who hits 40 HR could very well make up for his at bats lost to strikeouts.

7)On Base Plus Slugging (OBP+Slugging %): This is a stat that has risen in popularity this year, but I'm really not sure what it measures. It's just...a number. One more...

8)Total Bases (really big formula): this is probably #2 for be behind K Ratio. It doesn't favor towards power hitters, as a guy who steals bases or has speed to leg out extra bases can do just as well. But it also requires getting on base a lot and running smart once you get there.

This post was inspired by the evolution of stats in baseball. As I said above, the OBPS has seen a great rise in value (at least on Sportscenter :)) this year. If you go back 6-8 years, it was Runs Produced. Next year, it could be something else.



Mean Gene: "You know, I don't think it's a question - Goldberg, I don't think it's a question of who's next, I think it's a question of who's left?"
Goldberg: "No, see, that's where you're wrong. It ain't who's left, it's - WHO'S NEXT?"

"Just how hardcore am I? Well this morning, I drank milk that was two days past the expiration!"
-Norman Smiley

"She is one of them! She's CANADIAN!"
-Stevie Ray
Promote this thread!
Jubuki
Kolbasz








Since: 16.7.02

Since last post: 4310 days
Last activity: 4293 days
#2 Posted on
OPS is looked at a lot like an "effectiveness per AB" kind of measure. It's how many times one reaches base, and how many bases are being taken each time. However, I think it's something of a misleading number, as you're essentially giving the batter credit twice for hitting for a high average. Ichiro is a great player, but he's being spotted .720 in OPS, which can be a red herring for his overall effectiveness when the guy's only got 33 RBI (low even for a leadoff hitter). I'm not sure it's as solid a measure of offensive production as some treat it. I think I'd rather see something along the lines of:

(Walks + Steals + HBP + Total Bases)/Plate Appearances

Maybe you could even factor in a negative for double plays, which lessen the number of available chances for run scoring. Maybe call it Offense Produced. That way, you take into account as many factors as possible - speed, patience, power, good baserunning, knowing how to stay out of a double play, etc. Granted, I've always been a batting average and K/BB ratio nut myself, but I also think there's still some wiggle room for the sabremetricians out there to improve.

(edited by Jubuki on 28.7.02 1920)


Chris
The AIR RAID CRASH
drjayphd
Scrapple
Moderator








Since: 22.4.02
From: Long Island

Since last post: 3 days
Last activity: 3 days
AIM:  
ICQ:  
Y!:
#3 Posted on
Wasn't there some other stat ESPN came up with based on total bases and AB's? I forget the name or the formula, but I think it was something like "Batter Efficiency" or something. That one sounded good, but I'd go with OPS.

Besides, isn't counting triples in any stat more of a measurement of a player's speed and ballpark? I mean, Nixon hit a triple today, and it seemed like the only reason he could is because he's got fairly good speed, hits left-handed, and pulled the ball into that little nook in Fenway's RF. Some stadiums are just too small for anyone to hit a triple.
deadbeater
Morcilla








Since: 12.2.02
From: Parts unknown

Since last post: 4289 days
Last activity: 4289 days
#4 Posted on
My stat that I call Scoring Runs Ratio (Runs+RBI-HR/Total Team Runs Scored). For an everyday average player, it should be around 15-18%. For the three-sixth slots in batting order, or leadoff, it should be around 23-28%. If someone is 33% or above, he is extremely valuable to a team. Note percentage should total around 200. If the lineup is all 20-24%, like the Yankees are probably, the the team is well-balanced.



"Business has just picked up" says JR. Then right on cue, Molly enters the arena! Never was JR any better in introducing someone. --Raw 7/15
drjayphd
Scrapple
Moderator








Since: 22.4.02
From: Long Island

Since last post: 3 days
Last activity: 3 days
AIM:  
ICQ:  
Y!:
#5 Posted on
Yeah, that reminded me of a question I thought of a while ago... For the first two or three hitters in a lineup, shouldn't the three main stats be average, HR, and runs scored instead of average/HR/RBI? They're supposed to score the runs, the people behind them drive 'em in.
odessasteps
Scrapple








Since: 2.1.02
From: MD, USA

Since last post: 45 days
Last activity: 12 days
#6 Posted on | Instant Rating: 6.41

I'm an OPS guy, as a disciple of the Bill James/Rob Neyer cult.

Once I read that even hitters like Tony Gwynn said Batting Average is an overrated statistic, that was enough for me.
Guru Zim
SQL Dejection
Administrator








Since: 9.12.01
From: Bay City, OR

Since last post: 1 day
Last activity: 4 hours
AIM:  
#7 Posted on
The stat that I'd like to see would involve:

Steals + runs + Hits with men in scoring position + rbi + HR + walks

But I'm not sure how to weight everything in order to make it work. For example, is a walk equivalent to a steal? It's probably worth more. I value stealing 2nd above stealing 3rd - so it would be nice to account for that. I also feel that HR overinflate the value of a player in a lot of stats, but I'm not sure how to minimize this effect without undervalueuing the stat.

Hm...

I suppose I could use total bases, but the problem I want to avoid is ranking a .350 player that never gets home or brings in a run higher than a .250 guy that hits .400 in the clutch and steals a lot.



Your a retarted looser.
Tom Dean
Bockwurst








Since: 30.8.02
From: New York, NY

Since last post: 3224 days
Last activity: 2593 days
AIM:  
#8 Posted on
My thoughts:

- These stats that subtract HR from R and RBI are silly. A homerun is both a run scored and a run batted in. It should count as both. R+RBI would be a very reasonable stat.
- What does OPS measure? It measures the ability to get on base, and the ability to advance baserunners. These are the two basic aspects of offense (besides baserunning, which is the third, significantly less important factor). As you can see, it should count batting average twice, because singles both get people on base and advance runners. OPS does have its flaws. A .400 OBP/.350 SLG, in reality, is better than a .350 OBP/.400 SLG. OPS also is not park adjusted, and as mentioned, does not cover baserunning (or fielding.) Nonetheless, it is a good stat to use in conversation, newspaper columns, etc. because it's easy to figure out and is based off of stats that people already accept.
- If you are willing and able to move to stats that take a long time to explain, Bill James' runs created and Baseball Prospectus' EQA are good ones. The formulae for both are complex, although at least both end up with results that are on a scale people recognize. (Runs created look like runs scored or RBI totals, EQAs look like batting averages.) EQA is adjusted for home park and pretty much everything else you could ever think of. Runs created is not, but does have the advantage that it attempt to tell you how many runs the player added to an offense. This answers TheBucsFan's concern about what is really being measured... it's not only meaningful on its own scale.
- Whatever stat you use, it should be evaluated by how well it correlates with run scoring in real life. (i.e., use it on a whole lot of real life teams and see how accurately it ranks them.) You can't just make stuff up because it sounds neat or seems to describe how you suspect the game works.

(edited by T.R. on 28.7.02 1613)


"How YOU Doin'?"
- Tom Dean, weekly at [slash]
"History is being make-ed... somebody here is getting their head completely shaved off"
- David McLane, PPV opening promo

gonna build a giant drill and bore straight into hell releasing ancient demons from their sleep-forever spell so they can walk upon the earth and get recituated and run the diet pill pyramid that MC Pee Pants has created
Jubuki
Kolbasz








Since: 16.7.02

Since last post: 4310 days
Last activity: 4293 days
#9 Posted on
Well, that's exactly the reason WHY it's subtracted - the run batted in on a home run IS the run scored. So leaving it in place is a bit like OPS is now, counting one of the basic parts of the statistic twice for no good reason. Walks can advance runners and drive in runs, too, y'know.

(edited by Jubuki on 28.7.02 2115)


Chris
The AIR RAID CRASH
ges7184
Lap cheong








Since: 7.1.02
From: Birmingham, AL

Since last post: 89 days
Last activity: 12 days
#10 Posted on
How about a stat that would be a ratio of runs batted in compared to the number of runs that could have been batted it. For example, if no-one was on, you would have only the opportunity to knock in one run, yourself. With men, on first and second, you would have the opportunity to knock in three runs. Add up RBI's and opportunities and come up with a ratio: RBI's/opportunities.

Of course, this doesn't consider that getting yourself on base and into scoring position is of some value too. So maybe the ratio could become: Runs scored+RBI's/opportunities.

But then again, if you hit a double with no outs, but nobody can knock you in, that's not really your fault. So probably the real answer is that there is no one stat that can determine the value of the baseball player. However, maybe something like the above could be of some value in such evaluations.
Guru Zim
SQL Dejection
Administrator








Since: 9.12.01
From: Bay City, OR

Since last post: 1 day
Last activity: 4 hours
AIM:  
#11 Posted on

    You can't just make stuff up because it sounds neat or seems to describe how you suspect the game works.



I had you up until then. Unfortunately, I thinkk half the fun IS making up stats on how you suspect the game works, and then seeing how they fit.

For example, someone needs to make a stat up that can show the difference between between players who play on good teams and players that play on awful teams, but are identical in ability and production on an individual level.

Say for example you have identical twins who have identical abilities. This is just an excercise in thinking, don't get literal on me...

Player A is on Team A. Except for Player A, every other person on the team bats .250 and there are no power hitters. The team averages 6 HR per batting slot.

Player B is on Team B. Except for Player B, every other player on the team bats .330. The team averages 20 HR per batting position.

Player A and Player B play in parks that are statistically similar. They are both in the same division so they face the same pitching.

Ok. So, what will A and B's stats look like? A will probably be alone on base more often than B is because his team doesn't hit as well, so he might get more steals. They should both have equivalent batting averages and walk totals. A will probably have fewer runs and RBI because the team hits less than team B.

What stat is there that shows me that A and B have the same abilities and potentials? This is what I want to be able to find in a stat, not a measurement of how good the team surrounding a player was - but what you can expect from each individual.

OPS? This only works because it doesn't factor in RBI and Runs. Is this the only way to do it? Also, OPS doesn't really factor in power, which although it isn't very useful in this comparison, is an important stat, isn't it?



Your a retarted looser.
TheBucsFan
TheChiefsFan








Since: 2.1.02

Since last post: 3 days
Last activity: 1 day
#12 Posted on | Instant Rating: 4.42
Maybe something along the lines of (total bases/# times on base)/at bats would work. You'd wind up with a number like .005 as a good score, though, so some adjustment would have to be made.



Mean Gene: "You know, I don't think it's a question - Goldberg, I don't think it's a question of who's next, I think it's a question of who's left?"
Goldberg: "No, see, that's where you're wrong. It ain't who's left, it's - WHO'S NEXT?"

"Just how hardcore am I? Well this morning, I drank milk that was two days past the expiration!"
-Norman Smiley

"She is one of them! She's CANADIAN!"
-Stevie Ray
Guru Zim
SQL Dejection
Administrator








Since: 9.12.01
From: Bay City, OR

Since last post: 1 day
Last activity: 4 hours
AIM:  
#13 Posted on
I guess for me personally, the holy grail of baseball stats would let answer questions like the following:

Was Mickey Mantle as good as Ken Griffey Jr.?

And have some sort of non ballpark biased, non era biased, non backing team biased stat that could prove it.

(edited by Guru Zim on 28.7.02 2210)


Your a retarted looser.
pieman
As young as
he feels








Since: 11.12.01
From: China, Maine

Since last post: 2 days
Last activity: 12 hours
AIM:  
ICQ:  
Y!:
#14 Posted on | Instant Rating: 8.21
As odessasteps said up there, Bill James is the man.

Win Shares. Read it. It explains it all.

Win Shares is your stat.



He's Rolie Polie Olie - and in his world of curves and curls, he's the swellest kid around.
PhilRippa
Chourico








Since: 4.1.02
From: Fairfax, VA

Since last post: 1871 days
Last activity: 476 days
#15 Posted on

    Originally posted by pieman
    As odessasteps said up there, Bill James is the man.

    Win Shares. Read it. It explains it all.

    Win Shares is your stat.



Okay pieman - time to explain. Because no one I know can explain how they work. Even folks who are saber freaky.

I mean as of right now, I think I might have received one win share for a Yankee game that I attended back in 98.

Phil



Hate World, Revenge Soon
DVDVR.com
TheBucsFan
TheChiefsFan








Since: 2.1.02

Since last post: 3 days
Last activity: 1 day
#16 Posted on | Instant Rating: 4.42
I've never read the book win shares, but from how it was explained to me it seems to favor players on better teams.



Mean Gene: "You know, I don't think it's a question - Goldberg, I don't think it's a question of who's next, I think it's a question of who's left?"
Goldberg: "No, see, that's where you're wrong. It ain't who's left, it's - WHO'S NEXT?"

"Just how hardcore am I? Well this morning, I drank milk that was two days past the expiration!"
-Norman Smiley

"She is one of them! She's CANADIAN!"
-Stevie Ray
odessasteps
Scrapple








Since: 2.1.02
From: MD, USA

Since last post: 45 days
Last activity: 12 days
#17 Posted on | Instant Rating: 6.41

We had a thread on Win Shares a while ago, probably soon after the book came out.

I'd go back and look down below.

PhilRippa
Chourico








Since: 4.1.02
From: Fairfax, VA

Since last post: 1871 days
Last activity: 476 days
#18 Posted on

    Originally posted by odessasteps

    We had a thread on Win Shares a while ago, probably soon after the book came out.

    I'd go back and look down below.




Found it

Click Here

Of course - this still doesn't mean, I will understand it :-)

Phil



Hate World, Revenge Soon
DVDVR.com
odessasteps
Scrapple








Since: 2.1.02
From: MD, USA

Since last post: 45 days
Last activity: 12 days
#19 Posted on | Instant Rating: 6.41

Understood. I mean, the "Win Share" short version is like 12 or 13 steps and only approximates it.

(this is a little fresh in my mind, since I just reviewed both books for the new issue of the magazine.)

Tom Dean
Bockwurst








Since: 30.8.02
From: New York, NY

Since last post: 3224 days
Last activity: 2593 days
AIM:  
#20 Posted on
Jubuki: re: runs produced: Every other run is double-counted; why shouldn't homers be too? Scenario 1: Player A doubles, Player B doubles. The team scores one run. Two runs produced are given out: one run scored, plus one RBI. Scenario 2: Player A homers. The team scores one run. Why should two runs produced not be given out? One run is one run.

re: OPS: You're right that a walk can advance a runner. Obviously a lot less than a single, since it advances a maximum of one base, and sometimes (eg, man on second) it doesn't advance the runner at all. But, it certainly can, and that is part of the reason why OPS underrates OBP relative to SLG. OPS is not perfect, but it is way more accurate than the other simple stats that are easily available and understood.

Guru: You're right, I shouldn't have said that you CAN'T think about stats some other way. Baseball fans love stats almost by definition and love to tinker with them. But to do the kind of analysis that you're talking about, I think it's like any other kind of analysis, and you have to use the scientific method and start with the data rather than with the theory.

Win Shares is indeed the closest thing currently to the accounts-for-everything Big Stat you seek. We are already pretty good at measuring offense. The big advancement in Win Shares is that it appears to account for fielding far better than any other past system. Put together the fact that it measures offense well, and the fact that it measures defense well, and you have a pretty damn good start.

I do think it has some flaws. They mostly have to do with playing time. A pitcher from the 1970's who leads the league with 350 IP is given a lot more WS than a pitcher from the 2000's who leads the league with 250 IP, even if they pitched equally well per inning. James rates Clemens above Maddux for this reason. I am not sure this really makes sense. Also, you can basically get WS by just hanging out. If you played every day, you'd have to stink pretty badly to get single-digit WS. A lot of people have criticized the idea that you can get WS by playing worse than Joe Minor Leaguer could probably do.

It's not biased towards winning teams btw. Teams are given three times as many WS as their real-life wins. 100 wins = 300 WS, 60 wins = 180 WS. Then the system divides the shares among the players by, as people here have accurately said, a very complex process. If you had two equal superstars, the guy on the worse team has a smaller pool to draw from, but he is taking a higher percentage of that pool, so it comes out the same. If you think about it, it makes sense.

(BTW, James has Mantle #6 all-time and Griffey #73, and Griffey ain't exactly making up ground quick.)



"How YOU Doin'?"
- Tom Dean, weekly at [slash]
"History is being make-ed... somebody here is getting their head completely shaved off"
- David McLane, PPV opening promo

gonna build a giant drill and bore straight into hell releasing ancient demons from their sleep-forever spell so they can walk upon the earth and get recituated and run the diet pill pyramid that MC Pee Pants has created
Pages: 1 2 Next
Pages: 1 2 NextThread ahead: Strike date of AUGUST 16th now likely
Next thread: Basebrawling
Previous thread: Cards 10, Cubs 9
(1489 newer) Next thread | Previous thread
Shit shit shit shit shit. Perez is going to be a great pitcher. He's.... 21, 22? He's a lefty who throws too many pitches to get strikeouts... and he doesn't make it past the 5th inning. Today.
- Guru Zim, Pads Get Brian Giles... (2003)
The W - Baseball - Baseball StatisticsRegister and log in to post!

The W™ message board

ZimBoard
©2001-2014 Brothers Zim

This old hunk of junk rendered your page in 0.121 seconds.