Most of you who are baseball fans will already know what I am going to write in this article. I am not blazing a trail in any way with this column, but am just doing my duty to raise baseball literacy, so to speak. There are a few stats that are still frequently in use that need to go, not only as far as announcers are concerned, but as far as how you analyze what players are good, bad, etc. Some of them are misleading because they put the emphasis on the wrong things, some are just irrelevant, and some are fine stats just used in the completely wrong context. So put your prior notions of statistics aside for a second and hear me out.
On July 24th, Justin Masterson went 7 innings, gave up one earned run, and struck out 6 while walking one. On that same day, Tim Wakefield went 6.1 innings, allowing 7 earned runs, including a couple of bombs. Tim Wakefield got a win, Justin Masterson got a loss. In the All-Star Game this year in Arizona, Tyler Clippard recorded one out after giving up one hit. Technically speaking, he didn’t even record the out as the out was the result of Jose Bautista getting thrown out at home off a hit given up by Clippard. Clippard got the win for the game. The bottom line is that the pitcher has virtually zero impact on if he wins or loses a game. A pitcher winning is far more dependent on his offense’s production than anything else. In the NL, pitchers at least hit, so they at least have a crack at determining their W-L a tiny, tiny amount, but even then, most of the pitchers cannot hit at all and cannot even really help out on that level. I will at least listen to the idea that a pitcher working quickly and getting his guys back in the dugout quicker could very, very marginally boost the offense’s morale, but I do not think there is any factual evidence to back that up. Stop paying attention to W-L records for pitchers and you will be better at deciding if a pitcher is any good. If you just crave seeing a record next to a pitcher’s name, use Quality Start record, where you get a “win” for a QS (6IP, 3ER or less), and a “loss” for not having a QS. Frankly, that is not necessary, but if you crave a record, use that.
Errors (as a means for deciding who is a better fielder)
Imagine a player who literally cannot move. He makes every single play that he can get to but cannot make it to anything he has to move his feet for. His fielding percentage is 1.000%. Now imagine a guy who has a ton of range and can cover every single inch of the ground he is suppoed to cover, but some of the plays are really tough, since he is covering so much ground, and commits quite a few errors as a result. I do not think anyone would tell you the first guy is better. That is clearly an extreme example, but the point holds up in less extreme situations. The basic idea here is that making more plays is better than making fewer plays. Revolutionary, I know. Errors are clearly not a good thing, but they do not mean one player is worse than the other. Derek Jeter is the classic example of this. He has very limited range, and thus makes it to many fewer balls than others, but does not make a ton of errors so he is a well-thought of defender in some circles. Those circles are wrong. His lack of errors is not a result of his sterling defense as much as his limitations. Errors still matter, but you should not use them to evaluate who is a good defender. Oh, and do not even get me started on the Gold Gloves. The best way to win a gold glove? Hit the ball, defense be damned. But that is a different story. Errors can certainly indicate if a player makes mistakes, but really does not represent if one is a good fielder. UZR is a statistic designed to measure how often a player makes it to balls in his zone as compared to the average. A UZR of over 0 means the fielder is above average, and below 0 means below average. Easy enough. No one is going to get to every ball in his zone, but getting to more of them and making plays is certainly a good indicator of if a player is a good defender.
Let’s say a left-handed batter hits a grounder between the 1st and 2nd baseman. If they have a shift on, he is out. If the 2nd baseman is playing at double play depth, it is a hit. If the second baseman is a good fielder (see UZR), it may be an out, and inversely, if it is a bad fielder, it could be a hit. The bottom line is that a batter really has very little control over whether a batted ball will be a hit or not. Bill James, the father of sabermetrics and noted author, has even suggested that a batted ball’s chances of being a hit are almost completely random. Home Runs are a clear exception as fielders cannot catch those, and sure, a higher percentage of line drives would suggest a hitter is earning his batting average more than normal, but hitters, as well as pitchers for that matter, are largely at the mercy of luck as to whether or not a batted ball will be a hit. That is why pitchers who strike guys out are so valued. That is why batters who draw walks and hit home runs are so valued. Those outcomes are, in large part, devoid of luck.
It isn’t that batting average is useless, but just that you have to look at a million other things to determine if you should trust it or not, and even then it is a bit fluky. BABIP, Line Drive %, and a few other stats have to be consulted to even check if batting average MAY be helpful, so you are better off just using On Base Percentage and OPS (On Base + Slugging) to determine a batter’s worth as it weights extra base hits as being worth more than singles (which is accurate), and includes walks, which are often just as useful as hits btu without that element of luck. If you are feeling adventurous/not lazy, use OPS+ since it factors in the league average for you so you can compare instantly instead of having to look up other guys’ OPS and is already adjusted for differences in parks. You tell me which list looks more accurate of the best hitters in baseball.
Top 10 in OPS
- Bautista- 1.117 (which is ridiculous by the way)
- A. Gonzalez- .986
- M. Cabrera- .982
- M. Kemp- .981
- Berkman- .978
- Braun- .977
- Holliday- .974
- Fielder- .962
- Votto- .953
- Granderson- .943
Top 10 in BA
- A. Gonzalez- .357
- J. Reyes- .339
- M. Young- .334
- Kotchman- .328
- Braun- .324
- Votto- .324
- Bautista- .322
- Daniel Murphy- .319
- Kemp- .318
- V. Martinez- .317
With a stat so prone to fluky results, it should be no surprise that Casey Kotchman, Daniel Murphy, and Michael Young are all in the top 10 in batting average. They are still all good hitters, sporting solid OPS but they are not the 10 best hitters in the league by any stretch. The list above them is a much better approximation. Using OPS+ would be an even better approximation, but for our purposes, you get my point. Batting Average is not what you want to use to evaluate a player’s ability to hit.
ERA for Relievers
Let’s say there is a reliever who is the picture of consistency. He will always get you three outs, but will always give up one run. No more, no less, every time. His ERA would be 9.00. Would you still take him in your bullpen? He would be awful in some situations, especially tie games and one run games, but if you are up 2 going into the 9th, that is an automatic win. If you just need to keep them from putting a bit rally together in the 8th while up 3 or 4, he is ideal. He wouldn’t be the best reliever in my bullpen, but he absolutely would have a use on any roster. An ERA of 9 is horrible, but he is still a useful piece. By the same token, if you have a pitcher who goes long stretches being lights out, but then turns around and implodes every once in a while to the tune of a 9.00 ERA. That pitcher is far less reliable than the guy above, though they both have their ups and downs. Bullpen ERA as a whole is certainly helpful, but on an individual basis, it is not especially helpful. Since I know the Indians bullpen the best of MLB bullpens, let’s use theirs as an example. Chris Perez, the closer, has an ERA of 3.11 which is pretty good, but has a WHIP of 1.30 which is bad for a closer and absolutely life-threatening for any fans with heart conditions. He has a K:BB rate that is pretty close to 1:1 at 25:19. Those are not the stats of a good reliever. Tony Sipp has an ERA of 2.98, which is pretty close to Perez, but has a K:BB of 42:17, and a WHIP of 1.01. Pestano has an ERA of 3.12, but with a great K:BB rate (58:13) and WHIP (1.04). Sipp and Pestano have struggled lately, but for most of the year they have been much better relievers than Perez. WHIP and K:BB are the better stats to use with relievers because the main concern with a reliever is allowing baserunners to reach base and start rallies, and being able to get out of jams with Ks. The things we evaluate in relievers and starting pitchers are just different.
Frankly, I could turn this into a discussion about ERA vs FIP (fielding-independent pitching) and how ERA isn’t a great indicator to being with, and so on, but for simplicity’s sake, let’s just keep it to ERA being mediocre for assessing relievers. The basic point, however, is not so different from the argument against batting average since the defense behind a pitcher has a huge impact on how many runs are scored, etc.
The Eye Test
Watching baseball is a great way to get a feel for when a guy is playing well or not playing well, and a good way to tell if a guy is going to be a stud or suck, but it is not foolproof. I take great pride in my understanding of hitting and pitching mechanics and can sometimes tell if a guy is going to be a good major leaguer, but sometimes the eye test just does not work. Would you rather have a guy with a beautiful swing and an OBP of .320, or a guy with an ugly swing who gets on base at a .380 clip? No brainer. Ben Broussard had a gorgeous stroke. Only problem was that he sucked. The eye test can help you out on occasion, but when you have major league scouts constantly incorrectly evaluating a guy, it is clear that even the professionals need more than the eye test. Same goes for you. Use the eye test and see what you can see, but do not use it as a trump card for a player’s production.
I am going to stop things here and let you do some investigation on your own. Go look up ISO (isolated power- a measure of a player’s power), FIP (fielding-independent pitching for pitchers) ERA+, OPS+ and UZR/150. None are too difficult to understand and they will help you a lot in your attempts to decide if a player is actually good, just appears good, or something in between. Go forth and explore, and don’t be afraid of the new stats, just take them with a grain of salt and take into consideration what they are really measuring.
If you are really feeling crazy, look up “TINSTAAPP” or click THIS LINK to read a little about the Ubaldo trade as well as the idea that “There is no such thing as a pitching prospect”.