It is one of the biggest challenges in fantasy hockey: handling hot starts from unexpected players, or slumps from stars. How many games is enough? Will the trend continue? Should I cut bait?
Fortunately, analytics are great at answering these very questions. It is one thing to repeat the adage of buying low and selling high, but intuitively, anybody can figure that out. What is more valuable is knowing when to sell low or buy high. Those types of calls are tougher to make, and often require more temerity, but they are the types of moves that win championships.
With that said, here are some thoughts on early season surprises.
Alex Steen: Cemented alongside David Backes and T.J. Oshie, Alex Steen has notched a surprising league-leading 12 goals. However, he is an extremely strong sell if you have him. There are generally three things I look at in a case like this: on-ice shooting percentage, offensive zone start percentage, and quality of competition. Steen’s on-ice shooting percentage is 15.3% (strike one), his offensive zone start percentage is 40.5% (the most defensively-utilized trio on St. Louis: strike two), and his Corsi Rel QoC is an absurdly high 1.943 (strike three). His possession numbers are positive despite his usage, but it is difficult to expect anything close to this pace just based on his shooting luck alone, let alone the difficult assignments.
Kyle Okposo: Okposo is the beneficiary of the “third wheel” assignment alongside John Tavares and now Thomas Vanek, a role that gives him value on his own. Furthermore, his offensive usage is incredibly appealing, and his 10.5% on-ice shooting percentage is within the realm of sustainability because of the elite talent he is playing with. “Real” NHL analysis would call him sheltered with his cushy assignments, but for fantasy, we don’t care about that. Go get him.
Frans Nielsen: The untrained eye would call Okposo and Nielsen equals, but we are smarter than that here at Hockey Prospectus. The obvious difference: Nielsen is on the second unit alongside Pierre-Marc Bouchard and Josh Bailey. The more subtle ones: his five-on-five on-ice shooting percentage is nearly 16.0%, he is not a positive possession player, and his assignments are far tougher. He has some value because of his first unit power play role alongside Tavares and Vanek, but when the even strength production drops off, he becomes a fairly common commodity.
Tomas Hertl: I pumped Hertl up tirelessly in preseason based on the fact that he was slated to play alongside Joe Thornton and Brent Burns, and he cost very little. That said, nobody could have predicted his four-goal explosion (against the Rangers). He has been relatively quiet since then, but I still want him on my team. His possession rates remain absurd, and while we cannot quantify it, I suspect Brent Burns’ injury is partially to blame for his slump of late. The rookie wall is a concern, but you would be selling low if you moved him now, and he is still in a favorable environment.
Radim Vrbata: On the surprisingly offensively potent Phoenix Coyotes, Vrbata leads the charge with 12 points in 15 games. His even strength percentages indicate average luck (always a concern with a team-wide surge such as this one) but it is always a concern when half of a player’s output comes on the power play, as Vrbata’s has. He is shooting the puck at a very high rate, which is positive for fantasy, so I would call him a hold based on that factor. However, counting on the Phoenix power play has never been too advised, so keep tabs on that development.
Claude Giroux: Analytically, Giroux’s numbers have not changed all that much since his breakout 2011-12 season, outside of the fact that his competition has gotten slightly harder (as the Flyers rely on him for more and more key faceoffs). Still, he has been used more offensively since that season, and he leads the entire team in time on ice, so he is going to get opportunities to score. Like most of the Flyers, his shooting percentage luck has been poor, indicative of a rebound, but the concern is that he is not shooting the puck as much as he has been in past years.
Scott Hartnell: One of my favorite buy-low candidates in hockey to this point, because of his absurd combination of possession prowess and poor shooting luck. Go on, check it out for yourself. What you will see is a 35.0 Relative Corsi with a 1.8% on-ice shooting percentage. Yes, 1.8%! Many of the Flyers’ are suffering from poor luck, but none this bad. While the team’s power play has not helped boost his scoring totals, this seems too outrageous to be believed. There are 33 players (minimum five games played) who have had worse five-on-five luck. Most have negative Relative Corsi ratings; the next closest to Hartnell is Eric Tangradi at 21.0. There might be a chance he has been dropped outright in your league; I would make space for him if you could, because he always has provided a unique combination of power play time and penalty minutes/hits.
Eric Staal: The only thing that seems to be lacking for Staal is a dash of shooting luck, and perhaps proper expectations. Both his 100-point campaign in 2005-06 and his shortened 53 points in 48 games last year have perhaps obscured the fact that he is typically a 70-point player. Nothing to sneeze at, to be sure, but a notch under the elite fantasy level into which he has been tossed this season. You can expect an uptick in scoring, but of some concern is his recent pairing with his brother Jordan, suggesting more defensive responsibility. He typically does not provide auxiliary categories such as blocks or hits, either, and if your league counts plus-minus, he will surely hurt you there, due to the struggles in the Carolina net. If you can sell him low to someone who believes that he is a first round fantasy talent, I would do so.
Possession vs. shooting update
Clear trends emerging from this data as the season progresses (the blue line is close-game Fenwick, the red line is team shooting percentage in those situations):
1. Buy Kings! I don’t know how they do it every year, but look at that red spike downward on the left side of the chart: a 57% close-game Fenwick team shooting 5% in those situations is a massive red flag.
2. Watch the Ducks! Last year, the Ducks coupled their unusually high shooting with poor possession. This year, they appear to have righted the ship in the Fenwick department, but the high shooting is something to monitor if that tails off. One positive is that their shocking bad power play (last at 6.9%, 2.5% behind 29th Philadelphia) is bound to improve, helping their skaters in that regard.
3. Don’t buy Capitals, Oilers, and Leafs! In general, investing in teams with sub 45% possession over 10% shooting is a bad gamble. I won’t say to overbearingly “sell” these players, because it is early in the season, and sample sizes are quite small, but this is a negative trend to keep our eye on.
4. Movement: The Rangers and Islanders are creeping up the close-game Fenwick chart from the last time we observed the data on October 14. Washington has fallen off strongly.
Ryan Schwepfinger is managing editor of Hockey Prospectus and he was co-editor of the Hockey Prospectus 2013-14 annual. He is Media Relations Coordinator of the USHL’s Youngstown Phantoms.
Follow Ryan on Twitter at @ryanschwep.