December 6, 2016 by bsd987
As I’ve noted before, my statistics are weighted such that more recent events count more than older events. This has the effect of adjusting the ratings more to the form of the player, i.e. allowing the ratings to shift quicker when a player either finds or loses form. It doesn’t ignore older events per se—that would be too draconian, and probably incorrect: One or two bad matches shouldn’t erase a season’s worth of good, at least until proven otherwise. But a problem comes when a player completely disintegrates overnight.
That, at least right now, is the story of Michael Smith.
It’s not hard to pinpoint exactly when Michael Smith lost his form. Just two weekends before the Champions League, Smith averaged over 100 in his first two matches at the European Darts Trophy before narrowly losing a last-leg decider to James Wade in the quarter-final. All signs pointed towards Smith kicking on from an inconsistent but still successful summer. He had reached the final in Vienna, although he was never fully on song that weekend. But he had a good showing in Blackpool, dispatching Simon Whitlock in one of the best ties of the first round before getting knocked out by a resurgent Steve Beaton. Autumn seemed to be a new start for the Bully Boy.
Then came Players Championships 14 to 16 and the Champions League.
In Barnsley, Smith began to indicate things had gone amiss. Although he cashed all three days, he did so by beating out of sorts players. His six wins included only one win over an eventual Ally Pally qualifier and only two over players in the top 100. His losses included a pair of 6-1 defeats to John Henderson and Chris Dobey.
But it was in Cardiff that things went to hell. Smith was a non-factor, losing all three matches, and he was awful in doing so. His three-dart average was a mere 296.21, 14 points below his seasonal numbers. His set-up averaged declined nearly 10 points on his seasonal numbers as well, from just over 100 to 90.99. And he missed with three-in-hand at a double 6 times from 13 visits, when his seasonal numbers would have predicted just 3 such misses. A sharp decline in all three facets of his game saw him win just 18 out of 48 legs and crash out.
Smith hasn’t recovered.
Since that weekend, Smith has won one match—a 6-5 victory the very next tournament over Ted Evetts—while going out first match in eight tournaments and the Grand Slam qualifiers. Included in that run was a 6-0 whitewash defeat to Ken MacNeil and a 6-1 loss to Mark Walsh, neither of whom qualified for Ally Pally.
Overall, dating back to PC16, Smith has lost 14 out of his last 15 matches. Even Dean Winstanley has only lost 13 of his last 15 matches.
And that raises a statistical problem: Even though my statistics are designed to account for a loss of form, they can’t adjust when a player is playing so few legs that the old data so greatly predominates.
Michael Smith still is rated a 509 in my handicap rating system because during this time, he has only played four ranking stage matches and lost them all. During that time, he played a total of 43 legs, winning 17 and losing 26. By comparison, Smith played 52 legs in Vienna alone and 34 in the Matchplay. Even with the decay built in, his last three months are so small in the sample that he appears to be an above-average player.
He’s so bad right now that it can only have a minimal impact on his statistics.
So I’ve done something just for him: I’ve gone and calculated his statistics if we throw in the non-ranking events from the past few months. I’ve calculated Michael Smith’s statistics from the Champions League and World Series Finals to see what effect, if any, it has on his rating. The chart that follows shows his statistics.
|Ranking Stage Events||305.14||101.60||109.85||509.46||196.7|
|Incl. CL & WS||300.43||98.33||102.54||496.20||301.1|
|Only Since PC16||291.87||97.66||101.07||489.17||154.5|
If I keep my sample to just his ranking stage events, Smith has a very good 509.46 handicap rating. Among World Championship qualifiers with at least 100 legs in my database, Smith would rank just under Jelle Klaasen (511.88) and just ahead of Adrian Lewis (508.62). He also would be ahead of supposedly in-form players like Simon Whitlock (503.61), Darren Webster (501.71), and Alan Norris (495.80).
But once we add in both Champions League and the World Series Final, Smith’s ranking tanks. He goes to a below-par 496.20. That ranks him in-between Steve Beaton (496.29) and the aforementioned Alan Norris.
And if we completely throw out his performances that were before the Champions League, his rating goes down even further. Now, He’s rated a 489.17, less than a full point ahead of Joe Cullen and Jamie Caven, and lower than non-qualifiers David Pallett and Andy Smith.
Clearly, something is wrong.
And while he’s well below his seasonal numbers all across the board, the worst numbers are in his 9-dart average. Since PC16, Smith’s 9-dart average is a paltry 291.87. If this were his season-long numbers, it would rank him 47th out of 60 Ally Pally participants for which I have statistics, just behind John Michael (292.00), and a fraction of a point ahead of Dimitri van den Bergh (291.82). He’d be behind a fully out-of-sorts Brendan Dolan and an even more out-of-sorts Robert Thornton, and is only scoring better than two of the 32 seeded players (John Henderson and Jamie Caven).
By comparison, his seasonal numbers place him 19th in the field out of 60 players, and a near-average 17th of the 32 seeds.
His checkout numbers are also down, but remarkably, he’s still a bit above average during the last two months. But when you consider his scoring and set-up play is also down, it means he’s getting fewer and fewer chances to check out. And being that finishing has (a bit surprisingly) been the strongest part of his game this year, a regression towards average there only exacerbates the problems elsewhere.
Anyway, what’s the good news for Michael Smith? He has the ability to bounce back. Even with two-and-a-half terrible months, he’s still rated better on the year than the world #5. And his decline, while across the board, is still relatively minor. His scoring decrease is the equivalent of one dart in single 5 instead of single 20 per leg, or less than one more treble 20 every other leg. He has the quality to get back on track, if he has the belief. But if he plays like he has the past couple of months, he can’t get past Ricky Evans.
And as old events begin to drop off the rankings in the coming months, his rating will have no choice but to shift downward.