RADIO
If you have played daily fantasy baseball at all, you know that the debate if batter versus pitcher (BvP) statistics should be trusted is a hotly contested argument. There are several groups of players that will use this as their top tool under the assumption that if a professional athlete is comfortable he is more likely to perform. Those that oppose BvP tend to argue that a lack of sample size and other factors such as platoon advantage play into this being a flawed statistic.
Statistical Background
The basis for many baseball stats is the large sample size a year or career of a player can create. For example, per FanGraphs, a hitter's strikeout rate will stabilize around 60 plate appearances. At 170 plate appearances a home run rate will stabilize. It takes roughly 910 plate appearances for a hitter's batting average to find its stabilizing point (roughly a season and a half). When put into that context it is clear that any one player is likely to never reach even a normalized strikeout rate against any one individual pitcher.
Proponents of BvP tend to argue that statistics prove that an individual hitter--or pitcher, for that matter--will go into the plate appearance knowing tendencies or how to attack the opposition. The level of comfort or ability to see a pitcher is seen more in short sample sizes, but argued that a comfort level is not something that can simply be measured. Those who follow the logic of BvP submit that familiarity, although nothing you can specifically measure, is more a part of the daily game play as opposed to statistics that can be rooted in some anomalies such as wind or park factors.
The Bryce Harper Conundrum
Perhaps the most well-known BvP debate used happened on opening day as the Atlanta Braves Julio Teheran was set to face off against Bryce Harper and the Washington Nationals. For his career, Harper has owned Teheran to the tune of a .429 average, 15% home run rate, .515 on-base percentage (OBP) and an amazing 1.071 slugging (total bases divided by number of at-bats) percentage. That played out on Opening Day as Harper yet again took Teheran deep to fuel the fire of BvP supporters.
In the case of Harper, BvP detractors will point out that 33 career plate appearances pales in comparison to the 1,542 career plate appearances against right handing pitchers that show Harper as a player with a career batting average of .301, a .398 OBP and a slugging percentage of .562. Those numbers suggest that Harper simply dominates right handed pitchers regardless of who they are and his Teheran dominance is just an anomaly that will work its way back to the mean over time.
A Week-Long Litmus Test
To see if there was an undeniable edge to players with dominant BvP, I ran a week-long test over the past seven days. Included were players in the starting lineup with at least eight plate appearances against an individual pitcher. That seemed to be the number likely to prompt BvP backers to take notice in a particular match-up that day. From there the players were split up into two groups; one group consisted of players with a batting average of over .400 against a pitcher, while group two consisted of players with a batting average of .100 or lower. This also was a good baseline as the current batting average across MLB is currently around .247 which means these two groups were the same deviation away from the average.
The results of this week-long experiment included all at-bats for the individual that week. It could be argued that they skew the test results when playing daily fantasy (or season long for that matter) when you choose a player you are locked into his entire game, which can also prove to be positive or negative. Over the seven-day stretch there were 45 players in the under .100 average group and 64 in the group over .400. What the test showed was a bit surprising to say the least!
Players in the .100-and-under group hit a combined .229 (38/166), while those in the over-.400 group only managed to hit at a .200 (47/235) clip. The fact that both are actually under the season long league average of .247 is not entirely surprising, however the fact that hitters with the worst BvP in the league actually fared better over a week of games.
Along the same lines, of the 64 players measured in the higher group, only 10 players managed multi-hit games, while nine of 45 players in the under .100 group managed a multi-hit game. Considering both groups contained players in which they had the platoon advantage, played equally home and away in their meetings. As mentioned before this is a small sample size, however this test seems to bear out that no one player or pitcher has a distinct advantage simply due to previous meetings.
Conclusion
After a week of tests and looking at baselines for averages, it is fairly evident to stay that although a good tool, BvP should not be used on its own merits when selecting players in a daily match-up. What it does point to is likely a bigger picture that can be exploited whether it be platoon advantage of the hitter or possibly the ineffectiveness of a pitcher against a particular type of hitter. Along with that it is also important to note that in today’s MLB the limits on pitches and specialists in bullpens, a hitter may simply not get enough chances in a game to do a ton of damage against a specific pitcher.
