Who can watch TVF Pitchers

Which stats are suitable for evaluating pitchers?

Two weeks ago, in my article “The batting title needs an update”, I claimed that the batting average is not very meaningful for evaluating a hitter and can lead fans on the wrong track when trying to compare two players. In fact, there are many more possibilities than the methods described therein, but let's first take a closer look at evaluating pitchers before delving into the depths of hitter analysis.

The equivalent of the batting average, the partially erroneous stat of the hitter, is probably the earned run average for the pitchers. ERA is the standard value par excellence when looking at the seasonal performance of a pitcher and is usually the first statistic that appears in a pitcher's stat line (after the pitcher win-loss statistic, which is actually so unnecessary that I neither explain it I'll ever mention it again).

The ERA is calculated by dividing the number of earned runs by the number of pitched innings and then multiplying this value by the number 9, i.e. the number of innings in a regular game. The ERA then states how many runs a pitcher would collect, on average, in 9 innings.
The MLB-wide ERA average was 4.51 last year.

Problems with the earned run average

There are many problems when calculating the earned run average, for example the fact that only earned runs are included. You may be wondering what the difference between an "earned run" and an "unearned run" is. One speaks of the latter if a run would not have come about without the mistake of the defense, and thus also of the pitching. In the end, however, it is the scorer's job to decide whether this run was “earned” or “unearned”, which should be a first bad signal.

Other problems are the different ball parks and the team-dependent strength of the defense that a pitcher has behind him. Let us now consider these two problems more closely.

DIPS (Defense-Independent Pitching Statistics) tries to solve the latter problem. One statistic from the “DIPS” category is “FIP” (Fielding Independent Pitching). Since not every pitcher has an equally strong defense behind them, this defense is set here to an MLB average value and we see what the ERA of a pitcher would have looked like with this defense. In order to achieve this, one only looks at the outputs of a plate appearance, which the pitcher can influence directly. Home runs, strikeouts, walks and hit-by-pitches. The first three stats mentioned are often also known as “Three True Outcomes”, as they (with rare exceptions) do not involve defense.

A small example: A ball in the 5-6 hole (the zone between third baseman and shortstop) is relatively easily converted into a ground out by a good defense, but with a poor defense it can easily become an infield single.

If you watch MLB games often, you will at some point hear the terms “pitcher-friendly ball park” and “hitter-friendly ball park”. This is due to the fact that, unlike American football, there is no exact size of the playing field, but only minimum values ​​and values ​​recommended by the MLB. As the name suggests, some ball parks are particularly beneficial for pitchers and others for hitters.

A pitcher who has an ERA of 5.00 at Coors Field of the Colorado Rockies and another pitcher who has the same ERA at Petco Park of the San Diego Padres should not and cannot be equated. Coors Field and Petco Park are usually the classic examples of a hitter-friendly and pitcher-friendly ball park and will continue to be used by me in the future.

The ballpark problem is attempted to be solved by “ERA-”, among other things. A “Park Factor” is included, which is given to each stadium and then additionally (in addition to other calculations, see picture) divided by the average ERA value of the American / National League. The average value is 100, with a value of 125+ being very bad and 70 being an excellent value.

The adaptation to the ball park types is also possible for "FIP", but I do not want to go into more detail at first.

More statistics

Another important value is “BABIP” (Batting Average on Balls in Play). Glenn DePaul wrote in an article for Beyond The Boxscore:

"Explaining batting average on balls in play for pitchers still very well could be the holy grail of sabermetrics"

Glenn DePaul, "Projecting BABIP and Regression Towards the Mean"

For a detailed overview, I recommend reading his article, but I still try to explain “BABIP” as best as possible (let's see what comes out of it).

While “FIP” does without the balls that get into the field, these are the only things that count for “BABIP”. “BABIP” calculates the number of times a ball that is struck into the field of play becomes a base hit and can therefore be used by both pitchers and hitters.

The problem here is again in the defense, which has a different level for each team. Fangraphs also sees two other important factors in “BABIP”: luck and talent.

The former is actually self-explanatory, the same applies to the talent, since a good pitcher usually does not allow hard contact with the ball, which has a greater chance of becoming a base hit (for example: a weak hit vs. a hard hit in the 5-6 hole).

However, pitchers with a high “BABIP” are not automatically bad as a result, but rather this often indicates a bad defense or an outlier season. For a long time, BABIP usually aligns itself with the average, which is known as “regression toward the mean”.

This is also one reason why Voros McCracken (inventor of the “DIPS”) wrote in 2001, in an almost legendary article about DIPS at “BaseballProspectus”:

“There is little if any difference among major-league pitchers in their ability to prevent hits on balls hit in the field of play. [...] There is little correlation between what a pitcher does one year in the stat and what he will do the next. In other words, what Eric Milton‘S hits per balls in play was in 2000 tells us next to nothing about what it will be in 2001. This is not true in the other significant stats (walks, strikeouts, home runs)."

Voros McCracken, "Pitching and Defense: How Much Control Do Hurlers Have?"

A small example might help: Lance Lynn had a GDP of .319 in 2015, but only a GDP of .246 the next year (in this case 2017, since it failed in 2016). In the last two years, however, it rose to .336 and .322, just above his career average of .302. Everything indicates that his 2017 season was an outlier in this statistic. But he is not a bad pitcher with a career average of .302, after all, he was number 5 in the AL Cy Young voting last year.

“BABIP” should therefore also be viewed with caution and not so heavily included in the forecast of next year's ERA or general performance, especially if this is far below the average BABIP value of the pitcher. Nevertheless, this helps us a lot in the evaluation of a pitcher, because we see exactly in what percentage of the cases we can expect a hit ball to become a base hit (on average, of course).

WAS

One number that is probably best for assessing a player at a glance is probably “WAR” (wins-above-replacement). The term “WAR” has become almost a buzzword, is used for hitters and pitchers and has meanwhile also found its way into American football. Briefly explained, WAR indicates how many wins a player brings to the team, as opposed to any player who could be secured for a minimal deal.

In general, I don't just look at the WAR value to get a picture of a player, but also look at the individual other statistics to get an overall picture of a player.

WHIP

If you've read my last article about the Arizona Diamondbacks, you are already familiar with this term, but you may not know what it means. Quite simply, this indicates how many walks and hits a pitcher makes per inning. The calculation is made by dividing the number of walks and hits by the number of pitched innings. The evaluation should be intuitive: the lower the value, the better.

Starter vs. Reliever

There are two general groups of pitchers: Starting Pitchers and Relief Pitchers. The first group usually starts the games (there are now “openers” who pitch the first inning, but they are by no means used by every team and even those who use them do not do this in every game) and pitch as many innings as possible. Relievers are the pitchers who come into play after the starting pitchers.

With this distinction alone, it should be clear that these pitchers cannot be evaluated immediately. In my opinion, for example, ERA is an unnecessary value when comparing two relievers, as they often only come into play for one or two innings and then I don't really care how many runs they average in handing in nine innings (of course it shouldn't be 7.00+ and a reliever with an ERA below 3.00 is usually good too).

Another value that I don't pay too much attention to are saves. A save is given to the relief pitcher, who ends the game for the winning team without the team leading with more than three runs if the team takes over (there are a few mini-details that are initially uninteresting). For saves, the so-called “closer” often comes into play, which is usually the best reliever of a team.

Unfortunately, many managers still pay close attention to this stat and only want to use their closer in these situations. A bad example of this are the Baltimore Orioles and their closer Zack Britton. During the 2016 AL Wild Card Game against the Toronto Blue Jays, Buck Showalter (manager of the Orioles) passed on his best reliever, even though there were several situations in which he would have more than needed it.

Just because a pitcher has a lot of saves does not automatically mean that he is very good, it only means that he came into play in the right situations. If I pay attention to saves, then more to the “blown saves”, ie save situations that are not used.

When it comes to relievers, I pay special attention to their WHIP, i.e. their number of hits and walks per inning, and their strikeout-to-walk ratio, which should be self-explanatory. I also often take a look at the percentage of ground balls and how many flyballs become home runs (HR / FB%) (also with starting pitchers to see whether he is more of a groundball or flyball pitcher and how vulnerable he is for home runs is).

In general, you should never use just one of these statistics to evaluate pitchers (and also hitters), but rather the statistics mentioned above, other statistics and, above all, the game tape to get an actual picture of a pitcher can.