Frozen Tool Forensics: Consistent Fantasy Performers like Lindholm, Huberdeau, and Terry

Chris Kane


The name of the game this week has been trade deadline talk. Monday's deadline created an interesting weekend of news, takes, and of course one voided trade. We won't really be touching specifically on the deadline as the moves and implications have been very well covered on the site. Start here for a rundown on all that happened, plus links for analysis on the fantasy relevant trades. Go here for coverage specific to cap leagues. And go here, and here for some Ramblings coverage and fall out.

Today we are going to return, as we occasionally do, to players' ability to put up consistent production. In the past we have used some of this data to talk about streaky players in a given season, and trying to see if we can find streaky players over the course of multiple seasons. I wanted to think about it a little differently today. I am thinking about managers who are fighting to win their current weeks and are staring at a bunch of free agents and wondering who to add. We already take into account a player's general performance (point pace), deployment/time on ice etc. Sometimes those things feel a bit like a coin flip. What if a team needs a single-game Hail Mary? Maybe it would be helpful to know if a player tends to get multi-point games. Or maybe they see a two-game schedule ahead and want a player who is more consistently putting up points for the best shot at getting something.

All that and more this week on Frozen Tool Forensics: Most Consistent Players

In order to look at this I pulled data from Frozen Tool's Consistency report. We have looked at this in the past, but there are some improvements we should mention here too. It has the typical info about a player (position, team, etc.), but then breaks out into games played, stat-per-game, and consistency metrics for points, multi-point games, goals, and assists. Today we are going to focus on point consistency, and multi-point consistency.

If we were to just sort by general consistency (defined as the percent of games where a player gets a point) all we see are the league's top scorers. This makes all of the sense. Players who score the most are going to score in the most games. What we need to do then is account for a player's point pace. Using the data set we can generate a trend that will allow us to find what a player's expected consistency rate is given their points per game, and then whether or not their actual consistency rate is better or worse than that rate.

First up we are going to take the 'more consistent' group. The table below is the top ten (or at least started as the top ten, but since a bunch of players were all essentially tied it expanded).

NamePosTeamGPGP w/ PtPTSPTS/GPConsistency %ExpectedΔ