Offseason Priorities

[westofyou]

Batting average exists because the main rule of the game was for the pitcher to pitch to the bat. And the hitter got to declare where he wanted it thrown

There was no HBP, no walks, no sacs.

Just Hits, errors and hands outs (Outs recorded)

[Superdude]

Originally Posted by AtomicDumpling

Here is an interesting scenario:

Imagine that baseball had not been invented until the year 2000, well into the information age of computers. Now that there is so much data available about every game and so much processing power at our fingertips, do you think that Batting Average would ever have been invented as a statistic?

Batting average is basically a relic from the 1800's that became popular before records were kept on things like walks, errors and HBPs, and before extra base hits became common.

The flaws in battting average are so readily apparent by today's standards that I don't think batting average would ever have become popular if baseball had been invented in the modern world.

Of course it would. Regardless of who's right about it's value to run production, it's a fundamental part of hitting. Not every stat has to aim for total value to be a useful evaluating tool.

[AtomicDumpling]

Originally Posted by Kc61

Drew Stubbs fanned 166 times this year in 544 PAs. His K rate was 30.5%.

In my opinion, if he could have cut that number in half, made contact in another 83 at bats, he would have been a better hitter.

I guess you disagree, to each his own, but that's my view.

In some relatively few instances no-contact is better than contact.

Usually, contact is better.

I think everyone can agree that Drew Stubbs is a poor hitter. You can say it is because he strikes out too much. I can say it is because he makes too many outs and doesn't get enough extra-base hits. I think changing his swing to make more contact would improve his batting average, keep his OBP the same and make his SLG worse.

You might ask "if he makes more contact and improves his batting average why wouldn't that raise his OBP?" I would answer, because that approach would cause him to walk less and make weaker contact (weak contact means fewer base hits per hit ball than hard contact does). Changing his approach affects all his at-bats, not just the ones where he would have struck out using his old approach. You can't just isolate out the strikeouts and change your approach on those ABs. A change in approach affects every plate appearance. Maximizing contact results in weaker contact, so while he might snag a base hit where he may have struck out in the past, he might also make weak contact on a pitch he would have crushed in the past. Weak contact is less likely to get through the infield or become a base hit. So Stubbs would see some balls that would have been hits in the past that are now fielded for an out. Now consider also that making contact more frequently means you are less likely to walk (you can't walk if you hit the ball). More frequent but weaker contact can result in the same OBP as less frequent but harder contact. Weaker contact will definitely result in a much lower SLG than hard contact.

[*BaseClogger*]

Did this thread come out of the archives from 2003?

[westofyou]

Originally Posted by *BaseClogger*

Did this thread come out of the archives from 2003?

I found it among my Linear Weights notes and my Craig Wright scrapbook

[AtomicDumpling]

Originally Posted by Kc61

The best argument against BA is that base hits are influenced strongly by luck and other factors beyond the batter's control.

A base hit (notably singles and doubles) could easily be an out if the ball were hit slightly differently, or the fielder was positioned slightly differently, or the fielder was better, etc. This is a good argument.

The argument fails IMO when a hitter consistently gets lots of base hits per times at bat over a number of years. Some hitters have the ability to get hits for a variety of reasons.

On the other hand, when a batter has a high BA once or twice in a long career, then the stat may not mean as much.

Those are all valid points for certain.

I would add that the biggest argument against batting average is this: Simply getting a base hit is not the goal of a batter.

The goal of a batter is to put runs on the scoreboard for his team.

Base hits are not a good indicator of who the best run producers are. Going to the plate with the goal of getting a base hit is not a good approach. The batter needs to go to the plate with the goal of reaching base safely without making an out anywhere on the field (ie. don't cause a runner to be retired either) and once you reach base try to reach as far around the bases as you can.

Batting average does a very poor job of measuring a hitter's rate of success of achieving his goal -- OBP is much more accurate because it includes Walks and HBPs.

Batting average also does a poor job of measuring the hitter's degree of success at achieving his goal -- SLG is much more accurate because it factors how far the hitter got beyond first base on his hit.

OBP and SLG (and especially OPS and wOBA) much more accurately measure how effective a hitter was at achieving his goal: putting runs on the scoreboard for his team.

[Superdude]

Originally Posted by AtomicDumpling

I think everyone can agree that Drew Stubbs is a poor hitter. You can say it is because he strikes out too much. I can say it is because he makes too many outs and doesn't get enough extra-base hits.

He makes too many outs BECAUSE he strikes out 34% of his at bats.

[Kc61]

Originally Posted by AtomicDumpling

IWeak contact is less likely to get through the infield or become a base hit. So Stubbs would see some balls that would have been hits in the past that are now fielded for an out. Now consider also that making contact more frequently means you are less likely to walk (you can't walk if you hit the ball). More frequent but weaker contact can result in the same OBP as less frequent but harder contact. Weaker contact will definitely result in a much lower SLG than hard contact.

Good answer. Let's take a different hypothetical.

Take Drew Stubbs with his 166 Ks. Compare him to a different hitter, Batter X.

Batter X is identical to Stubbs in every way, same approach, same results, except he makes more contact. He strikes out only 83 times and makes contact and hits a fair ball an additional 83 times.

IMO, Batter X will have better overall numbers than Drew. IMO, an identical hitter making more contact will do better.

I'd prefer Batter X as a hitter to Drew Stubbs. It's kind of obvious but feel free to disagree. I'm moving on to other things.

[AtomicDumpling]

Originally Posted by Kc61

Good answer. Let's take a different hypothetical.

Take Drew Stubbs with his 166 Ks. Compare him to a different hitter, Batter X.

Batter X is identical to Stubbs in every way, same approach, same results, but he makes more contact. He strikes out only 83 times and makes contact and hits a fair ball an additional 83 times.

IMO, Batter X will have better overall numbers than Drew. IMO, an identical hitter making more contact will do better.

That's my point. More contact is generally better, all other things being equal.

Those two bolded items don't compute. If the results are the same he will have the same overall numbers.

If the batter you propose existed he would have better OBP and SLG than Drew Stubbs if some of those 83 extra contacts were not outs, that would be a better hitter than Drew Stubbs. The batting average doesn't change anything.

You are making the circular assertion that if Drew Stubbs were a better hitter he would be a better hitter. Obviously, if he hit the ball better then he would have better BA/OBP/SLG stats.

I often hear the notion that if he would strike out less he would be a better hitter. OK fine. How is he going to strike out less? How will that affect the rest of his at-bats? If it really were that simple why hasn't he done it already? People have been saying that about hitters for decades but it simply doesn't work that way.

You can't isolate out the strikeouts from the total package. That is like saying "if Drew Stubbs didn't hit as many routine bouncers to the shortstop he would have a better batting average." How is that different than saying "If Drew Stubbs didn't strike out as much he would have a better batting average"?

[_Sir_Charles_]

Originally Posted by Superdude

I have a hard time with this concept. I get that OBP/SLG is more important, but if given the choice between .200/.330/.450 and .300/.330/.450, it makes no difference? BA may not correlate perfectly with runs scored, but to say it means nothing just sounds like oversimplification to me.

The problem I have with this is that it's looking at ONE PLAYER instead of the group of a TEAM. The .200/.330/.450 guy may not be producing more or less than the .300/.330/.450 guy on his own, but the additional hits are increasing the odds for those following players on the team to contribute. Moving defenders, creating holes, etc. Sometimes the sabermetric guys make it seem like a walk and a single are basically the same thing..."he's avoiding an out"...but for me there's a VAST difference in the two. A hitter who has good contact skills does more than just have a higher BA. He fouls off more pitches and works the pitchers pitch count higher, he's less likely to strike out, he's more likely to advance runners with sacs, hit & runs, etc. It's not just ONE guy and his ONE stat line. It's the interaction between all the pieces of the puzzle that sabermetrics seems to overlook quite a bit. I think KC is spot on here. Overall, the team has pretty good slugging, but VERY poor contact skills. Filling those missing skills in the team overall is only going to help.

[_Sir_Charles_]

Originally Posted by AtomicDumpling

I would add that the biggest argument against batting average is this: Simply getting a base hit is not the goal of a batter.

The goal of a batter is to put runs on the scoreboard for his team.

Could not disagree more.

The goal of the TEAM is to score runs. The goal of the individual is to do his part to make that possible. The goal of the individual is to get on base with a hit, walk...whatever...or do what he can to make it easier for another teammate to score a run...advance a runner with an out, work the count and force a pitching change, etc.

Looking at a single player overlooks the fact that this is a TEAM game. They're all interconnected.

[AtomicDumpling]

Originally Posted by _Sir_Charles_

The problem I have with this is that it's looking at ONE PLAYER instead of the group of a TEAM. The .200/.330/.450 guy may not be producing more or less than the .300/.330/.450 guy on his own, but the additional hits are increasing the odds for those following players on the team to contribute. Moving defenders, creating holes, etc. Sometimes the sabermetric guys make it seem like a walk and a single are basically the same thing..."he's avoiding an out"...but for me there's a VAST difference in the two. A hitter who has good contact skills does more than just have a higher BA. He fouls off more pitches and works the pitchers pitch count higher, he's less likely to strike out, he's more likely to advance runners with sacs, hit & runs, etc. It's not just ONE guy and his ONE stat line. It's the interaction between all the pieces of the puzzle that sabermetrics seems to overlook quite a bit. I think KC is spot on here. Overall, the team has pretty good slugging, but VERY poor contact skills. Filling those missing skills in the team overall is only going to help.

On the contrary, the saber guys are the ones that ARE factoring in BA as a part of the whole team's offensive contributions. That is exactly what the whole idea of correlating each stat with real-world run scoring does. These correlations are done on team stats, not individual stats. Team Batting Average simply does not correlate well with run scoring in real MLB games. OBP and SLG correlate much better with run scoring on a team basis than BA does. wOBA and OPS are even better.

You are absolutely correct that you have to evaluate the big picture, where a player's contributions at the plate echo far beyond his own individual at-bats. I just think the saber approach does a far better job of that than batting average does.

Saber guys don't consider a walk and a single as basically the same thing. The run value of a single is .47 runs for his team. The run value of a walk is .33 runs for his team. So you can see that sabermetrically, a single is quite a bit more valuable than a walk. The run value of a home run is 1.41 runs, so a home run is a lot more valuable than either a single or a walk. Sabermetrics captures all this at exactly the ratios they exhibit themselves in real MLB games. Batting average would have counted those singles and home runs exactly the same as each other and would have totally ignored the walks.

Players who hit the ball into play see on average fewer pitches than hitters who strike out or walk. You can hit the first pitch, but you can't strike out or walk on the first pitch. It takes a lot of pitches to strike out or walk.

Strikeouts are no more harmful than other outs on average. It is statistically proven, as I showed earlier in this thread.

The benefit gained by advancing runners with sacs and hit-and-runs etc is cancelled out by the double-plays and fielder's choices over the course of a season.

[Superdude]

Originally Posted by AtomicDumpling

On the contrary, the saber guys are the ones that ARE factoring in BA as a part of the whole team's offensive contributions. That is exactly what the whole idea of correlating each stat with real-world run scoring does. These correlations are done on team stats, not individual stats. Team Batting Average simply does not correlate well with run scoring in real MLB games. OBP and SLG correlate much better with run scoring on a team basis than BA does. wOBA and OPS are even better.

My argument is that trying to correlate batting average to runs scored is a faulty method in the first place. It's obviously not going to correlate much because it's only one piece of a much bigger puzzle. It's correlation to runs scored has very little to do with it's utility IMO.

Originally Posted by AtomicDumpling

Batting average would have counted those singles and home runs exactly the same as each other and would have totally ignored the walks.

What's the buzzword around here, straw man? No is arguing this point, and once again, I think you're making the false assumption that because batting average doesn't tell everything, it has no value at all.

[AtomicDumpling]

Originally Posted by _Sir_Charles_

The problem I have with this is that it's looking at ONE PLAYER instead of the group of a TEAM. The .200/.330/.450 guy may not be producing more or less than the .300/.330/.450 guy on his own, but the additional hits are increasing the odds for those following players on the team to contribute. Moving defenders, creating holes, etc. Sometimes the sabermetric guys make it seem like a walk and a single are basically the same thing..."he's avoiding an out"...but for me there's a VAST difference in the two. A hitter who has good contact skills does more than just have a higher BA. He fouls off more pitches and works the pitchers pitch count higher, he's less likely to strike out, he's more likely to advance runners with sacs, hit & runs, etc. It's not just ONE guy and his ONE stat line. It's the interaction between all the pieces of the puzzle that sabermetrics seems to overlook quite a bit. I think KC is spot on here. Overall, the team has pretty good slugging, but VERY poor contact skills. Filling those missing skills in the team overall is only going to help.

Originally Posted by AtomicDumpling

Saber guys don't consider a walk and a single as basically the same thing. The run value of a single is .47 runs for his team. The run value of a walk is .33 runs for his team. So you can see that sabermetrically, a single is quite a bit more valuable than a walk. The run value of a home run is 1.41 runs, so a home run is a lot more valuable than either a single or a walk. Sabermetrics captures all this at exactly the ratios they exhibit themselves in real MLB games. Batting average would have counted those singles and home runs exactly the same as each other and would have totally ignored the walks.

Originally Posted by Superdude

What's the buzzword around here, straw man? No is arguing this point, and once again, I think you're making the false assumption that because batting average doesn't tell everything, it has no value at all.

There is an example of someone arguing precisely that point just a couple posts ago and that is the post I was replying to at the time.

[VR]

The problem is, over the course of a season, all the sabr stats even out. (I love sabr)

In a one game, or three game series, they mean very very little.