Don’t Call It a Comeback
Jim Furyk and Steve Stricker are on a pretty good run right now. It’s not a total surprise either. They spent a good chunk of last year racking up points atop leaderboards, but then they took some time off to recover from injuries. Figuring out when to hop back on the bandwagon with guys like them isn’t easy: How do you know when they’re back to their old selves? And is there anything we can quantify in the interim to guess exactly how much penalty they should receive, if any, when they’re in the middle of their comeback?
I tried answering this question with a methodology similar to the one I used for rookies and the volume index. I assigned each player a ‘recency index,’ which reflects a weighting of how much each player’s results came from recent events versus far-back events. If a player has most of his long-term log coming from past results, he will have a low recency index. This is an indirect way of capturing the players who have extended layoffs, which are proxies for time off due to injury.
So how does recency index correlate with relative performance? Here’s how the data looks:
This one requires a bit of explanation, since recency index is a little more complicated than a simple volume index. If a golfer played a tournament every week during the 75-week window, he would have a total recency index just under 0.5, which is why you see the relative performance close to zero at that mark. That represents a ‘standard’ distribution of results. A player with a very high recency index will have the majority of his results from the last 10 weeks with no long-term backlog — a rookie on the Tour, for instance — and his penalty will come primarily from the low volume index, which is overlapping here.
Injuries, Layoffs, and Performance
The meat of what we’re trying to get to — players with extended layoffs — comes from the left side of this graph. And the penalty for extended layoffs is on par with the low-volume rookie penalty, if not worse. It’s tough to know exactly what a low recency index is picking up that accounts for the penalty. Maybe it’s residual injury that hasn’t healed. Or maybe it’s rust from being away from the game for so long. Either way, I don’t think that figuring out exactly why is all that important. As long as we can measure it and it improves our long-term predictions, that’s good enough for me.
This is why I’ve been similarly bearish on Furyk this year during his recovery tour. It’s tempting to hop on the bandwagon before prices go back up, but . . .
A) PGA pricing doesn’t really work like that, and
B) There’s a predictable performance penalty that goes along with recovery.
You can always say, “Stats be damned, I think he’s back,” but you had better be certain enough to overcome the long-term trend in that graph.
