Fantasy Chronicles: Points Above Replacement

When I last talked about fantasy baseball, I focused on the best way to draft catchers. Essentially, I realized that the few offensively-talented catchers were sandwiched between superior players, so it would make most sense to draft the best players available. The goal of course is to draft a team with the highest projected points that I can possibly get, so passing up on better players didn’t make sense. I looked at the “net rank” for catchers, which is how they ranked in projected points minus how they ranked in ADP. This approach made the most sense to me because if I was drafting a catcher with the best net ranking, I knew I wouldn’t be hurting my team by drafting them in their ADP with better players potentially available.

This made sense but over time, it started driving me a little mad with a flaw it had. This chart that displays the 2017 fantasy season among position players, starting pitchers, and relief pitchers should explain why:

There are a few takeaways from this graph. One: position players typically are worth more than pitchers. Two: Starters are worth more than relievers. Three: There are more total valuable position players than there are pitchers. And four: There’s a pretty steep drop off in talent among pitchers after around the 20^th best pitcher or so.

How does this change the way we should draft? Well going with the strategy I had before, after the first couple rounds, the player with the most projected points was going to always be a position player. Drafting the best available player every time was going to be a problem because after drafting a bunch of nice position players, only scrubs for pitchers would be left for me to draft. Theoretically, to once again achieve the most points possible with my team, it would make most sense to draft the elite pitchers while I still could, especially knowing that the sheer volume of valuable position players would give me a safety net of being able to wait to pick them up later.

To test out this new strategy, I needed a number besides total points to quantify their real value. I know that a 300 point position player is ideally worth the same as a 300 point pitcher. However, that pitcher has to be more valued because we know a pitcher of his caliber is more rare than a position player of that caliber. With a suggestion from a friend, that brings in “points above replacement.” How many points is this player worth compared to the points of what the replacement-level of that position would be? Replacement level can be tricky to define. Since there are nine pitchers on a team in the twelve-team league I’m in, a replacement pitcher would be however many points the 108^th best pitcher (12 x 9) is projected to get, which is in fact 186 points. Position players, as expected, have a higher replacement level at 258 points.

Here’s the top 10 players by points above replacement (PAR). As expected, more pitchers appear on this list than they would have if it were just ranked by points. This is only the top ten of a very, very large table but this pattern continues throughout the top 50 for PAR. I made two separate columns for points above replacement. The first one (PAR) is for PAR for each position, differentiating types of position players and using separate replacement levels for each position. The second one (PAR*) is more as originally described, just using the replacement level for all position players overall. In the end, the difference between the two was never too significant but I could see a situation when drafting for a specific position that hasn’t been filled out yet on the roster when one is more appropriate than the other.

Like I’ve been emphasizing, the goal is to figure out ways to maximize your team’s total points. Is PAR the best strategy? Even throughout mock drafts, I tested out different strategies and the PAR-strategy worked out the best every time. The method works best of course if other people aren’t following it but overall most people in most leagues are going to look at just projected points and ADP. Previously, seeing catchers like Posey and Sanchez get drafted super early didn’t make sense but really the high demand for them reflects their PAR. For whatever reason, and maybe other leagues have methods to reduce the large drop off in pitching talent, this didn’t get reflected in the ADP of pitchers. For example, ESPN’s top 10 has no pitchers, yet six of the top ten by PAR were pitchers. Four of my first five draft picks ended up being pitchers and I still ended up with a pretty good offense. I think my team will be just fine.