yyys123
Geregistreerd op: 08 Jul 2019
Berichten: 945
|
 Geplaatst: 12-07-2019 07:11:03 Onderwerp: d think that New Jersey was a |
 |
|
|
| One of the e-mails I receive frequently about the application of advanced statistics in professional hockey regards “PDO”, and why the metric is so often referenced when discussing outlier performance. PDO is nothing more than a combination of shooting percentage and save percentage, expressed in thousands. It’s a simple calculation, but imperative when conducting analysis and forecasting future outcomes. The theory behind PDO is that shooting percentage is primarily luck-driven, and save-percentage is primarily luck-driven, and at the team-level, teams will consistently regress towards this 1,000 (i.e., the league average) number. Teams with extremely high PDO’s, say 1020 and above, are great bets to regress unfavorably. Teams with extremely low PDO’s, say 980 and below, are great bets to regress favorably. From time to time, we’ll see small deviations in genuinely great and genuinely terrible teams. But in most cases, it simply pays to (a) be skeptical that any percentage-fueled run is real; (b) focus on winning the shot-differential battle, because shot-differentials will predict future outcome far better than past shooting and save percentages will. PDO was at the heart of the 2013-2014 Toronto Maple Leafs debate – a team whose predictable and catastrophic end-of-year collapse pushed professional hockey into the analytics era. A bunch of smart hires were made by organizations around the league, and it seemed as though the debate over percentage-fueled runs and team-level shot quality myths were put to bed. Still, there seems to be some lingering doubt. Many, many words have been spilled about the 2013-2014 Colorado Avalanche, a team that – despite endless precaution – decided to double-down on mythological shot quality, ignoring innumerable red flags in the process. It wasn’t just the Avalanche organization buying stock, either. Bovada, an online sportsbook with a vested interest in outcomes, opened with Colorado as a 98.5 point team. On the other hand, that same online sportsbook opened up with the New Jersey Devils as an 83.5 point team – 15-points less than Colorado. Are these two teams fifteen points different? It’s possible the answer is yes, but not in the way you’d think. First, let’s look at each team’s ability to control play via Corsi%, starting with game one of last season and running it through today’s data. We’ll use a 10-game rolling average to smooth out results. Not a whole lot has changed from last year to this year, which is signified by the vertical line at the game 82 mark. New Jersey has consistently earned a better percentage of the shot-share, never once dipping below the 50% threshold over any 10-game stretch. Colorado, on the other hand, has been consistently subpar at controlling play. Other than a five-game window (31-36), they’ve been regularly under 50%. If you looked solely at the possession numbers and were aware of the tight correlation between controlling the puck and winning in today’s NHL, you would think that New Jersey was a playoff caliber team. Colorado? A lottery team. But, the hockey gods are funny sometimes. We know Colorado’s off to a horrendous and predictable 3-6-5 start, but the possession numbers don’t explain why things suddenly went south. Nor does it explain why New Jersey – who was a possession world-beater last year – failed to make the post-season. So, let’s go to the percentages, captured by the aforementioned PDO. Again, it’s more or less a measure of “puck luck”, and the likelihood of a team’s number regressing to 1,000 is extremely strong. We’ll roll Colorado and New Jersey’s PDO over 10-games to again smooth things a bit. Colorado sat well above the 1,000 mark for the vast majority of last season. New Jersey sat well below the 1,000 mark for the vast majority of last season. Whereas Colorado (8.07% Sh%, .931 SV%) saw all of the bounces at 5-on-5, New Jersey (7.12 Sh%, .914 SV%) did not. I think the dividing vertical lines on both of these graphs are amazing in the sense that they capture precisely what we’re looking for in terms of forecasting future outcome. When it came to a team’s ability to control play at 5-on-5 via Corsi%, both teams in 2014-2015 are reasonably near their respective 2013-2014 performance. This is because puck possession is repeatable. On the PDO graph, it’s the total opposite. The shooting and save percentages have flipped entirely, which is consistent with what we have seen in PDO volatility across many different teams over many, many years. New Jersey may have made the right move going from Martin Brodeur to Cory Schneider, but a goaltending switch wouldn’t explain how the team jumped from 26th to 14th in shooting percentage seemingly overnight. Randomness, of course, would. Combine that with generally out-possessing the opposition, and you have a respectable 6-4-2 record. On the Colorado side, the team has seen somewhat unfavorable percentage luck, but it’s far closer to the league averages than anything the team experienced last year. And, of course, the team is still getting drilled in the shot department. It’s a combination that generally ends up in fan bases paying attention to the draft lottery, rather than preparing for the post-season. Hydro Flask Kopen . Bryant, who signed a five-year, $34 million contract as a free agent with Cleveland in March, reported symptoms on Monday morning, a team spokesman said. Hydro Flask Korting . -- Former Penn State coach Joe Paterno was admitted to the hospital Friday for observation due to minor complications from cancer treatments, his family said. http://www.belgiehydroflask.com/ . Catch all the action on TSN starting at 10:30pm et/7:30pm pt. Toronto won at Denver and Utah, but lost in Portland and Sacramento. The Kings loss was the most recent game for the Raptors. Hydro Flask Belgie . Many of those eyes are in the United States of America, a country that the sport has wanted on its side for some time. On Sunday the football Gods delivered once again, only this time it was time for good old USA to experience as much heartache as joy. Hydro Flask 40 OZ Belgie . It will be their 15th head-to-head meeting and fourth in the post-season (Sunday at 2pm et/11am pt on CTV) and for his part, Brady isnt downplaying just how big the game is for him. "Im excited - Its everything you could ask for as an athlete," the New England Patriots quarterback told WEEI Radio in Boston on Monday.TSN.ca has unveiled its initial list of free agents and the talent available would have any general manager with cap space salivating. The list of course, will look a lot less robust once all the re-signings are final and the franchise tags are handed out. Names like TJ Ward and Jairus Byrd arent likely to ever see the open market. And Jimmy Graham, he isnt going anywhere. Still there are several names that will become available to the highest bidder come March 11.dddddddddddd Receivers like Eric Decker, Hakeem Nicks, and Anquan Boldin, pass rushers like Greg Hardy, Michael Bennett, and Jared Allen, and cornerbacks like Alterraun Verner and Brent Grimes could find themselves in different jerseys soon after the deadline opens. Cheap Jerseys China NFL Jerseys Cheap NBA Jerseys Wholesale NHL Jerseys Wholesale MLB Jerseys Cheap Soccer Jerseys China Wholesale NCAA Jerseys Wholesale Football Jerseys Wholesale Basketball Jerseys Wholesale Baseball Jerseys ' ' ' |
|
