Frozen Tool Forensics: First Half Success Stories
Chris Kane
2022-09-09
A few weeks ago on Frozen Tool Forensics, we focused on potential breakout players using a breakout threshold process outlined elsewhere on the site. For the full break down check out the article. Part of the analysis was looking at players that had strong second half splits. That got me thinking generally about first half/second half splits and that is where we are going to be for this week's article.
This week on Frozen Tool Forensics: 2021 First Half Success Stories
Generally, when we talk about splits (players performing better in one section of the season vs another) we encounter a lot of noise. For most players seeing them perform better in the first half than the second half or vice versa is often variance driven rather than some kind of repeatable skill. This variance could be as simple as shooting luck, or maybe we have a player adjusting to a new place in the lineup, or they get injured, or they get to take an injured player's spot. Through it all though we do occasionally have some players who do seem to have notorious slow starts (eh hem Kevin Fiala).
For the purposes of this article, we are going to be diving into some of the players with the more dramatic splits from the 21-22 season. Seeing their name listed shouldn't necessarily be a red flag that they will repeat in the 22-23 season. Giving players a label as a first half or second half performer would require a bit deeper of a dive that we won't get to today.
Preamble out of the way so let's get to the process. To start we are using the same methodology as in the article mentioned above. I ran two Big Board reports (available in our Frozen Tools), with the split date of January 15th. I exported those reports to be able to combine and isolate the columns I am most interested in. I also converted everything into points per game to account for the fluctuations there.
Just for completion’s sake, there are 303 players who performed better in the first half than in the second, while there are 319 players who performed better in the second. That left about 185 players who were the same. If we account for small sample sizes, say played a total of 15 games both in the first and second halves we have 225 players who played better in the first half, and 222 in the second.
The focus of this article though is going to be about players who performed better in the first half. I am mostly concerned about players who put up a fantasy relevant point pace, and players who saw a reasonably impactful (a 20+ full-season point pace) swing between the halves. That brings us down to about 62 players. I further removed players who it was specifically reported were playing through injuries in the second half (i.e. Mark Stone), or who were provided unsustainable opportunities to start the season (i.e. Evan Rodrigues).
Here is our top ten:
Name | Age | Pos | Team | Per Game PTS 1st Half | Per Game PTS 2nd Half | Δ in Splits | 1st Half GP | 2nd Half GP |
NAZEM KADRI | 31 | C | CGY | 1.5 | 1 | 0.50 | 32 | 39 |
JESSE PULJUJÄRVI | 24 | R | EDM | 0.8 | 0.3 | 0.50 | 33 | 32 |
ALEX OVECHKIN | 36 | L | WSH | 1.4 | 0.9 | 0.50 | 38 | 39 |
BRAD MARCHAND | 34 | L | BOS | 1.4 | 0.9 | 0.50 | 30 | 40 |
DEVON TOEWS | 28 | D | COL | 1.1 | 0.7 | 0.40 | 24 | 42 |
ANTHONY DUCLAIR | 26 | R | FLA | 1 | 0.6 | 0.40 | 30 | 44 |
ALEX KILLORN | 32 | R | T.B | 0.9 | 0.5 | 0.40 | 40 | 42 |
SONNY MILANO | 26 | L | 0.7 | 0.3 | 0.40 | 35 | 31 | |
TIMO MEIER | 25 | L | S.J | 1.2 | 0.8 | 0.40 | 34 | 43 |
LEON DRAISAITL | 26 | L | EDM | 1.5 | 1.2 | 0.30 | 35 | 45 |
I am not sure it is necessary to talk about Nazem Kadri at this point. His first half point pace was incredible, and not sustainable, and it has been addressed quite a bit. Jesse Puljujärvi though is an interesting case. He started the season with points in six straight games (nine points total), and 23 points in his first 28 games (and 67 full season point pace). The wheels fell off after that with only 13 points the rest of his season (a 30-point full season pace). There is one very clear correlation: time on ice. During the first half of the season, he was seeing between one and a half and two and half minutes of power-play time on average, and over 17 minutes of total time on ice. Through the second half he was all but invisible on the power-play and well down to 14.5-15.5 minutes of total ice time on average. A probably not super coincidental factor is also that Evander Kane joined the team at the end of January and likely played a significant role in changing Puljujärvi's deployment particularly on the power-play. With Kane back in the fold for 22-23 that first half flash from Puljujärvi seems definitely less likely.
Neither Puljujärvi or Devon Toews have a history of being first half performers, but unlike Puljujärvi, Toews didn't dramatically lose time in the second half. His split differences are a much more run of the mill variance. He was over a point per game through the first half, and that is just incredibly hard for a defenseman to maintain, particularly a defenseman who isn't even the top d-man on his own team. The only real flag for Toews in his shooting percentage (over eight percent is very high for a defensemen), but he has historically been a good shooter, plus he increased his time on ice, and his shots per game over his 20-21 season. He might not be a point per game for the first half again, but a strong season is very much in the cards for Toews.
Anthony Duclair is another one who doesn't really have a strong history of first half performances. In fact, his 20-21 season was exactly the opposite and it wasn't until the second half that he took off (with an increase in power-play time and better linemates). Duclair had incredible shooting luck to start the season, so like with Toews we are really looking at a variance driven splits. Duclair is very line dependent in Florida. He has been a pretty good fit alongside Aleksander Barkov, but with big changes coming in the Florida top nine (Matthew Tkachuk in, Jonathan Huberdeau, and Mason Marchment out – also missing Claude Giroux) his spot isn't guaranteed and that placement will drive his production much more than any recent split season data points.
That is all for this week
Do your part to support organizations working to make hockey for everyone.