Tyler vs Taylor Revisited

The draft is the father of the team

Something happens when you cross the wrong side of forty. Memories develop an aura about them. The past seems more significant. Part of this significance is nostalgia. But patterns also start to emerge from life events that previously went undetected. As Wordsworth put it: "The child is the father of the man." On the wrong side of forty you see that child more clearly than ever but you're powerless to guide him.

And so we come to another draft. Another number one pick. And another saviour. And bituman is happy. He told all of his friends in a fit of rage (when there was still snow) that it was his dream that Peter Chiarelli should become the General Manager of the Oilers. At the time he was not thinking McDavid, and even when McDavid happened, his spirit recoiled in disbelief.

As the McDavid moment arrives bituman can't help but think of the draft of five years ago. The draft is the father of the team. The Taylor vs Tyler drama is still somewhat fresh in our memories, and for the most part we've been pleased, haven't we? But some of us may have started to become less than pleased after reading the latest NHL re-draft; Taylor Hall dropped to fifth position (the re-draft is the father of the post). Thanks to Czechboy for bringing this to our attention.

Ben's verdict: Tyler Seguin

I decided to start by digging through the Copper and Blue archives, looking for clues. Could it be true? Is Tyler really that much better than Taylor? Most of the analyses and stats from back in the day called it as it actually happened, picking Hall as the right choice for our team.

Derek Zona fronted the interesting idea that since a lot of folks didn't figure that there was much difference between the first and second overall pick that the Oilers should trade the pick to Boston and squeeze a little more juice out of the lemon. But back in the day it seems like the one C&B author who might have nailed it, despite the laudanum, was Benjamin Massey:

Is there a chance that, in trying to avoid Eric Lindros, I'm picking Pat Falloon? I don't think so. Tyler Seguin is legit and everyone knows it. But I'd rather have a guy who's an A-grade player for seventy games a year for fifteen seasons than a guy who's an A+ for forty games a year in seven seasons… Ben's verdict: Tyler Seguin

True Lies

Ben has been correct with his prophecy that Seguin would play more games per season than Hall. Over the past five years Hall has played in 299 regular season games and Seguin 354. 55 games is a big difference. It's  not the 40 games per that the Massey decried, but there appears to be a significant difference in playing style and injuries between the two players.

As a result of playing more games Seguin has 189 even strength points to Hall's 168 and has out scored Hall 87 to 58.

When measured per hour Seguin produces more goals and more points than Hall, although Hall produces more assists. It seems the sometimes centre Seguin, initially billed for his play making skill, may be more of a pure scorer than Hall after all. The table below uses five-on-five regular season data from war-on-ice.

So on the surface, Seguin wins. But what if we look at the advanced stats?

Corsi Against

The above chart shows Expected (ECA60, horizontal) versus Observed (CA60, vertical) Corsi Against per 60 minutes. By hovering over the bubbles you can see the relevant cartesian coordinates. Expected Corsi is a theoretical number from Stephen Burtch (you can read more about in my previous post). It tells you what Corsi we might expected from Taylor and Tyler given a host of contextual variables including age, time on ice, zone starts, quality of competition, and quality of teammates.

You might notice that Hall's bubbles are farther the right. Obviously, he's played on teams that are defensively weaker than Tyler. Hall's Bubbles are also higher meaning that we observed that he was on the ice for more Corsis against. This is all as we might expect.

Bubble size in the above chart corresponds to the difference between ECA60 and CA60 (dCA60). Bigger bubbles in this case (Corsi Against) are not a good thing. Bigger bubbles mean that the player gave up more Corsis than expected.

What the above chart tells us is that when it comes to defense, Tyler's better.

Corsi For

The above chart shows Expected (ECF60, horizontal) versus Observed (CF60, vertical) Corsi For per 60 minutes. Corsi For corresponds to a player's ability to drive possession and shot attempts.

Hall's bubbles are lower and farther to the left. This is what we might expect. Tyler's played on better scoring teams than Taylor. Taylor's Expected and Observed Corsis are lower.

Bigger bubbles in this chart are good. Bigger bubbles mean the player in question exceeded expectations by producing more Corsis than expected. It's sort of more even here, although we can say for certain that Hall's been hurting the last two seasons under Eakins. For whatever reason he hasn't been driving possession.

dCorsi

This third chart is the combination of the previous two. It shows cobserved and expected Corsi-For Percentage per 60 minutes. Tyler's situation has been more positive. Burtch's formula expects him to have a higher CF% and he does, for the most part. Bubble size corresponds to dCorsi, or delta Corsi: the differential between Expected and Observed Corsi for percentage. Bigger bubbles are better. Although it's somewhat even, Tyler wins.

The above bar chart shows dCorsi only (same data as bubble size in chart three).

Over the past five seasons Tyler Seguin has an average dCorsi of 2.74, while Taylor Hall's average dCorsi 0.83. Seguin exceeds expectations to a greater degree than Hall.

Is the NHL.com re-draft right? You tell me. The post is the father of the comment.