A Brief Guide of Sabermetrics

Don't know where to start with sabermetrics? Start here!

In the ever-growing world of baseball, teams are trying to find new ways that allow them to acquire valuable players cheaply that can help them win games. Teams send scouts all around the U.S and even in international states so that they can hope to gain the next superstar of baseball. However, the number of prospects in the baseball community is so vast; it would take an eternity to decide which players to select or sign (if not drafted). To help sort which players are the best for the current team, sabermetrics is the primary tool used to evaluate these players. Sabermetrics is merely using numbers and percentages to assess multiple aspects of baseball and the athletes.

Although founded around the 1900s, it was Billy Beane and his “Money-Ball” technique that lead MLB into the sabermetric revolution. Today, there are sabermetrics departments in every baseball club, helping the team evaluate which players are doing good and which need to be traded. Although this topic is relatively new, it has grown into one of the most important aspects of baseball. For those baseball fans who want to be a baseball nerd, sabermetrics is vital, so in this article, I am going to go step-by-step to give you the general idea of what sabermetrics is.

What is sabermetrics?

Sabermetrics can generally be defined as collecting data from a hitter or pitcher and using that data to create numbers that assesses players. Another definition comes from Bill James, the godfather of sabermetrics, who describes it as "the search for objective knowledge about baseball." Data can range from hitting to base running, to fielding, and to pitching. Sabermetrics also encompasses data on player’s performances at specific conditions (runners at 1st with 2 outs) which can help us see which players are better or worse at certain times. The central philosophy for sabermetrics is a question and an answer. Sabermetricians ask questions such as “How can we create a metric that measures the power of a hitter?” and as a result of these question, new metrics are being made and used to evaluate a hitter. The most important thing to know about sabermetrics is whether a player can help a team win games by either scoring more runs or giving up fewer runs. The whole point of baseball is to win, so by measuring the contributions of whether a player can help a team or not can determine their value and reputation among the baseball fans.

Why sabermetrics?

There are many arguments said to be here, but the best reason for sabermetrics is that it gives us an objective view of how a baseball player is performing. If we evaluate the player based on pure human observation, the player might be overrated and underrated. Some people might disagree on how good a player is, and no one knows how good or not the player is. By using numbers to describe a player’s ability, everyone has an objective baseline to base their assumptions on a player. With more metrics being made on a player, there is more data to be used on who is a better baseball player or not.

Also, using many metrics is essential because, with more parameters, it will give us a clearer picture of how good a player is or not. Like choosing a restaurant when you go out to eat, you don’t just use one source of data to select a restaurant. You look at multiple websites and see which ones are the generally considered the best.

A brief summary of important metrics:

WAR: WAR or Wins Above Replacement measures the overall contribution of all of a player’s attributes (hitting, fielding, base running, and pitching) to how many more games a team will win. If the WAR of a player is 5.5, then the player contributes to 5.5 more wins for his team.

OPS+: Most baseball fans know that OPS is the combination of on-base percentage and slugging percentage which is a better metric to use than AVG. However, we can use OPS to find out how much better the player is doing compared to the average hitter including park factors. This is what OPS+ is. It measures how much percent the player is doing better or worse compared to the league average on a scale based on a median of 100. For example, if I have a player with 150+ OPS, he is doing 50% better than the average player including ballparks factors

ISO: ISO or Isolated Power measures the player’s ability to hit extra base hits or the player’s power. It favors doubles, triples, and home runs more than singles and the more extra base hits you get, the more valuable the player is to the team (since doubles, triples, and home runs are more valuable to a team for runs) and the higher the ISO is. It is usually calculated by SLG - AVG and .140 is the average. Excellent is over .250 and below .80 is bad.

wOBA: Generally considered one of the best hitting statistics in baseball, wOBA uses a different formula than OPS in order to solve. Unlike OPS, it uses specific run value for each hit value based on Linear Weights (another topic for later) and measures the exact hitting value of a hitter. As a result, it is a much more specific and better statistic to use to measure a hitter. wOBA is generally read the same as OBP, so anything between .300 and .350 is good and anything above .380 is a great player. Sabermetrics generally prefer to use this stat than OPS because there are flaws with OPS and wOBA is much more accurate.

wRC+: Just as like OPS+, wRC+ measures the offensive production of a player based on wOBA and park factors and puts them all in a statistic on how much percent better or worse the player is doing compared to the league average. Just like OPS+, the median is 100 and if I have a player with a 150 wRC+, he is doing 50% better than the average including park factors.

Let us use Aaron Judge and Jose Altuve from 2017 season as an example for the hitter metrics we described above.

Using Fangraphs, here is a table of the different stats mentioned above and a comparison between both of the players.

Seeing how both of these players were front-runners for the 2017 AL MVP award, we can use statistics to see which of these players is the better hitter.

As we can see, Aaron Judge has more power that Jose Altuve, seeing that he has 52 HRs and a .627 SLG. Besides, he has a high walk percentage that leads to a high OBP. However, he tends to strike out much more and has a lower BABIP than Jose Altuve.

On the other hand, Jose Altuve sacks some power for contact, seeing that he has a .346 AVG and a much lower SLG and ISO percentage. He tends to walk less, but he gets much fewer strikeouts than Aaron Judge.

Seeing that they contribute to their team differently (power/contact), we can use other statistics like the ones mentioned above to find which hitter is better and adds to his team.

As we can see, Aaron Judge has a higher WAR than Jose Altuve ( 8.2 vs. 7.6 WAR) and a higher ISO (.343 vs .202). He also has a higher wOBA (.430 vs .405) and a higher wRC+ (173 vs. 160). Jose Altuve has a lower strikeout percentage than Aaron Judge, but Aaron Judge has a higher walk percentage.

If Aaron Judge outperforms Jose Altuve in every category I discussed above, then why did he not win the MVP award last year?

The answer is straightforward. Aaron Judge was not very consistent in 2017, as he regressed in performance during the June/July and especially August. He also struggled greatly with scoring runners in scoring position. Jose Altuve, on the other hand, performed very well throughout the season and had seen great consistency through all of this hitting statistics no matter the conditions. Although controversial, I choose Altuve due to a better consistency. Although Aaron Judge has excellent hitting statistics, he regressed a ton during the mid-late season and did not play at the level he did during the beginning of the season. When considering an MVP, it is crucial for me to decide whether a player can perform exceptionally well throughout the whole season and Jose Altuve fits that criteria much more than Aaron Judge.

---

FIP: The primary goal of a pitcher is to limit the number of runs, but when using statistics such as ERA, some factors can change the number of runs scored such as lousy defense or luck from the balls being hit in play. FIP or Fielding Independent Pitching eliminates these multiple variables on defense and measures the average amount of runs a pitcher would allow by only using outcomes determined by a pitcher only (strikeouts, home runs, and walks), so we can see just how good the pitcher is on his own. FIP is read the same as ERA (2-3.3 is good).

xFIP: xFIP, although similar to FIP, is the expected Fielding Independent Pitching of a pitcher based on his past performances. This statistic gives a prediction on how good this pitcher will be based on outcomes the pitcher makes (home runs, walks, and strikeouts) and how many fly balls they allow. By using these statistics, xFIP measures the expected ERA of a pitcher based on BABIP (balls put into play) and the average fly ball/ home run ratio. We can see how good the player would be if he had home runs hit regularly against him (league average) based on how many fly balls he gives up and also how many walks and strikeouts he gives up.

ERA- or FIP-: Like OPS+ and wRC+, they measure how good the player is compared to the league average and how much percent they are better. For ERA-, it uses ERA, the park factor, and the relative ERA in the American or National League (depending on where the player plays) and FIPs uses FIPS and the relatives FIPS in the American or National league (depending on where the player play). The scale is also read the same, but vice versa. So, if I have an ERA- of 185, then the pitcher is doing 85% worse than the average.

SIERA: Skill-Interactive ERA or SIERA is the newest of the ERA predictors and is similar to FIP, but it is more accurate. It measures more accurately of a pitcher’s ability to throw well and perform well in the future. Like wOBA and OPS, SIERA is better than FIP, but the gap between both is not as big. Instead of limiting the equation to only outcomes the pitcher can make such as strikeouts, walks, and home runs, the equation answers “why pitchers are more successful at limiting hits and preventing runs.” The statistic is like FIP (in terms of measuring what it is) and shows that strikeouts are more important, walks are not bad if limited, and that more ground balls mean more outs.

K% and BB%: This is very self-explanatory, but K% and BB% measures the percentage of hitters that get a strikeout (K) or a walk (BB) against that pitcher. Although it is good on its own, it can be used in combination with other metrics such as FIP and SIERA to really measure how good the pitcher will do in the future and also how good the pitcher is on his own (since he controls how many strikeouts and walks he has).

Other important things to know: K/9, BB/9, HR/9, and WHIP

---

Let us use compare Max Scherzer and Chris Sale from last season using the metrics mentioned above.

Here is a FANGRAPHS table comparing the two pitchers during the 2017 baseball season.

These two pitchers, although compete in different leagues, have risen to be some of the best pitchers in the MLB. Both pitchers have started in the 2017 MLB All-Star game and the 2018 Game, so we should analyze both players to see how well they dominate the league.

When we first see the graph, we can see that they post a tremendous Win-Loss record, have very stellar ERAs and also post high WAR for their teams. However, we want to look at the statistics that matter, the “sabermetric” statistics.

When we analyze the graph further, we can see that Chris Sale has a much lower FIP that Max Scherzer, meaning that concerning “raw” skill, Chris Sale is much better than Max Scherzer. His ERA is higher than his FIP, meaning that he either has some bad luck going into games or has a below average defense to work with. We can also look at xFIP and see how well the pitchers will do in the future based on their “raw” skill. As we can see, Chris Sale has a much lower xFIP than Max Scherzer and would be considered a better pitcher since he would be predicted to give up fewer runs.

Chris Sale strikes out more people than Max Scherzer and also giver up fewer walks. Walks and Strikeouts are a significant factor when calculating for FIP and xFIP so Chris Sale would be better in that regard.

Other Statistics:

ERA-: 47 ERA- (Chris Sale) > 56 ERA- (Max Scherzer)

SIERA: 2.36 (Chris Sale) vs 2.73 (Max Scherzer)

K% and BB%: 37.6% K% and 6% BB% (Chris Sale) vs 34.4% K% and 6.2% BB% (Max Scherzer)

As we can see above, Chris Sale beats Max Scherzer in almost every aspect of statistics mentioned above, so if a person would be choosing between an high-caliber pitcher for their fantasy team, they would likely select Chris sale over Max Scherzer.

---

Of course, there are more statistics that can be used to evaluate a player, but these stats are the most frequent ones used. It is always better to have multiple stats to evaluate a player because no one stat is perfect and can evaluate a player in all aspects. By getting more stats about a player, there is an objective baseline to base the player on and whether he is a valuable hitter or not. In the next posts, I will explain more in detail about these statistics and what it considered good or not.

If you want to research more about sabermetrics on your own, I would highly recommend these 2 websites to look at, https://www.fangraphs.com/ or https://www.baseball-reference.com/ . These are some great websites for archives on a player and also advanced metrics (like the ones discussed here).

Created at August 1, 2018. Published 04/03/20