Our mailbags recently have been filled with mailbag questions, but last week we got a question that was so good that it’s getting an entire post!

So here’s a question for you all. Tage has 23 goals on an xGoals of 14, according to JFresh model. Does this mean he’s been lucky and regression is coming, or can it mean skills growth?

The Koz, Expected Buffalo Discord

To start, it’s nice to be able to check an xG vs actual goals scenario where a player is scoring more than expected after previously going through this exercise with Taylor Hall and Victor Olofsson.

Tage Thompson’s development track since joining the Sabres has been foggy. After the disappointing first season, he had positive growth starting the season in Rochester but only played a single game in 2019-20 after being injured in that lone game. From the end of 2019 to the end of the 2020-21 season, his on-ice numbers at 5-on-5 improved. Natural Stat Trick’s xG calculation has Thompson’s percentages improving year-over-year (Again, one game in 19-20).

Another piece to keep in mind about that 2021 season is that Thompson was able to put up improved numbers even with a half a season under Ralph Kruger where he saw extended stretches of games with limited ice time and a mismatch of linemates. His rate stats, however, did continue to improve year over year everywhere except for on-ice corsi and fenwick for, which shouldn’t be that surprising given the fourth-line minutes he was playing for a lot of last season on a team that finished dead last in the NHL.

The jump in goal scoring is the biggest story for Thompson’s season, as he finds himself at 23 goals through 52 games after only 18 goals in his first 145 games. A good first place to look for this jump comes from where the Sabres are generating shots with Thompson on the ice at 5v5, and in a refreshing change from other versions of Micah’s heat maps, there is a concentrated high volume of net-from shots that has increased from 2021 to 2022 and is a massive improvement from the black hole in front of the net in 2019.

Thompson himself has typically been pretty good about taking even-strength shots in the middle of the ice, and his individual xG/60 has decreased from 2021 to 2022 because of the higher volume of further shots, which, when you are your team’s leader in scoring and offensive production, comes with the territory that you get to take more “hero” shots for volume’s sake.

So now to transition to checking on the sustainability of Thompson’s new production, let’s start with looking at the distribution of their simulations. Last season, Thompson only finished two goals short of his expectation from Moneypuck of 10 – Something that would’ve been expected in 14.47% of hypothetical seasons. The finishing total of 8 would’ve been expected about 11.44% of the time. Half of the simulations would lead to an expectation of 8-12 goals.

Goal Total	Instances In The 10,000 Simulations	Percent of The 10,000 Simulations
6	478	4.78%
7	877	8.77%
8	1144	11.44%
9	1374	13.74%
10	1447	14.47%
11	1333	13.33%
12	1011	10.11%
13	811	8.11%
14	532	5.32%

In 2022, Tage has jumped from results under expectation to those that are, as Koz initially pointed out, much higher than expectation. Moneypuck’s calculation expects an extra goal compared to JFresh, but that is picking at straws there as that approximates to less than a percent difference in simulation frequency. Notably here, Thompson’s 23 goals only occurred in 1.22% of the simulations, and he only reached or eclipsed that number in about 3% of the simulations, and folks, we have ourselves an outlier.

(Might as well show the whole table of sim results for this one due to the outlier)

Goal Total	Instances In The 10,000 Simulations	Percent of The 10,000 Simulations
4	3	0.03%
5	4	0.04%
6	15	0.15%
7	50	0.5%
8	97	0.97%
9	193	1.93%
10	330	3.3%
11	538	5.38%
12	702	7.02%
13	944	9.44%
14	1024	10.24%
15	1111	11.11%
16	1100	11%
17	995	9.95%
18	903	9.03%
19	683	6.83%
20	457	4.57%
21	346	3.46%
22	204	2.04%
23	122	1.22%
24	81	0.81%
25	49	0.49%
26	20	0.2%
27	15	0.15%
28	6	0.06%
29	6	0.06%
30	1	0.01%
33	1	0.01%

There are still 26 games left this season for Thompson to continue putting up shots, so of course, the expectation here is that his pace will start to return closer to expectation. 15 expected goals in 56 games would put him around 6 more expected goals for the season if he stays healthy, which would place him about one short of 30. Of course, there is the consideration of good/bad shooting talent where Moneypuck acknowledges that some players are simply better or worse at shooting compared to what their model defines as the probability of a shot going in. Let’s use that 33 goal mark as my hope for Thompson’s final goal total on the year, as it will push him a little closer to expectation while not removing any shooting talent that may be considered.

We don’t have anything too reliable (yet) to drill too far into whether or not Thompson’s shooting talent is not better than that of the average NHLer being considered in the expected goal sample across the league’s shots, but there’s one quick piece that has been under our nose the whole time: Accuracy.

Compared to the first handful of seasons of his career, Thompson’s shooting percentage has doubled this year from about 5% to 10%. Of that 5% difference, almost all of it is addressed from a decrease in shots that are saved from almost 67% to almost 63%, while the percentage of shots he’s taken that have missed the net has dropped off about a single percentage point. 52 games would be too big of a sample in my opinion to attribute this increase of times beating the goalie to just luck, so I think in his age 24 season, Thompson is establishing himself as someone who can exceed the expectations of an NHL shooter on his shooting talent. Will that carry on for the rest of his career, one in which he will likely be settling into a first or second-line center role for the remainder of his current contract with the Sabres (Expecting a star trade or stud draft hit in the next couple of years) and potentially beyond? We’ll just have to keep watching to find out (Crowd boos cheesy ending).

Thompson’s breakout this year has been atop the list of positive signs for the future of the Sabres, and fans such as The Koz should take comfort in the expectation that he should continue to be a top player on the team for the near future that should be able to keep growing as he gets older and his teammates get stronger around him.

Also, join the xB subscriber Discord if you haven’t already because it could lead to the glitz and glamor of inspiring a future post!

Photo credit: Photo Credit: Jeanine Leech/Icon Sportswire via Getty Images

Data via: Natural Stat Trick, Moneypuck, JFresh, and Micah McCurdy