|The Power of Numbers|
Ryan Ma digs deep into the numbers, prepping you for the second half. A must read.
Back from a bit of a hiatus from Fantasy Hockey and “real-life” catching up with me. Those who have followed my columns throughout the years, pretty much know what I’m all about, Numbers and Data! In this day and age, data is everywhere. I would probably even argue that all it’s essentially in every aspect of our lives. Heck, being a school teacher as my “day job”, I’m constantly bombarded with data driven criterion to hit. In order to become an “effective” educator you need to do this with data, to better your teaching look at your student’s data, you need to use data to “know” your students… So if data is everywhere, and Fantasy Hockey is largely driven by it (number of goals scored, how many shots did player A take, what’s the GAA of the goalie), then why do some poolies continue to ignore the importance of data? I still observe plenty of poolies on the forums who go by their “gut feel” or other random non-data related justifications (“skating mechanics”, “prior draft lineage”, “trust me I watch a lot of their games” or “they’re my favourite player”), the interesting thing with that is 300 years ago we would have called that witchcraft, in this day and age we anoint it as “justifiable opinion”. Now don’t get me wrong, I value differentiated opinions just as much the next guy, but what I want to see is tangible substance-based evidence to justify a point rather than someone who is out there just to make noise. I’m hopefully here to do the former rather than the later. Buckle up and enjoy the ride!
A couple of weeks ago I created a thread on the forums about first-quarter numbers. I proposed how numbers in Fantasy Hockey generally fit a certain model and how you can take advantage of those numbers in order to help you win. This is the second instalment of the series and we’ll take a look back at my first quarter predictions and next week we’ll give you an insight into my predictions for the second half of this campaign, but first here’s a quick recap.
Bullet points from the thread (if you don’t have the time to sift through it.)
- Based on my research during the last 6-7 years, I've basically narrowed point production for all players down to 3 key factors which is, time on ice, PP time on ice and shots on goal.
- Those 3 factors are basically the most important factors in terms of correlation to point production, as with all cases, there's always going to be outliers (highly talented rookies who are handcuffed), pass first players (Jumbo Joe) and veterans who are being "managed" (Selanne, Iggy and Alfredsson) being the most significant, so the data doesn't always fit 100% of all fantasy players.
- With forwards, the strongest correlation to points is PP TOI which has a correlation co-efficient of a whopping 0.87, which is huge! Second is SOG with a correlation co-efficient of 0.85 and third is overall TOI with a correlation co-efficient of a whopping 0.84. (In the revision of 1st half numbers, those numbers are now 0.89, 0.88 and 0.87 respectively, which indicates an even stronger correlation)
- With defenseman the strongest correlation to points is SOG which has a correlation co-efficient of a whopping 0.80, which is not as strong as the forward stats, but good enough in our scenario. Second is PP TOI with a correlation co-efficient of 0.78. (In the revision of 1st half numbers, those numbers are now 0.88 and 0.84 respectively, which once again indicates an even stronger correlation)
Using FrozenPool (a tool that I can’t speak more highly of), I analysed the three factors which has the greatest effect on point production and looked at players who were over and under achieving based on those factors at the time. I earmarked and labelled a few of these players and here are the results from the data that I analysed. Green were players that I tabbed who were due for an increase in production, yellow and red were players who I identified for a decrease in production.
A few hits from this list, Giroux, Staal, Voracek and Simmonds were gimme picks, so it wasn’t at all surprising to really see them pick up their game during the second quarter.
Vanek’s boost could have been attributed to the trade from Buffalo to New York, which was probably expected. Brouwer was a big hit given the PP responsibility that he is given! Higgins saw a nice boost, but probably would have been nicer if he saw just a couple more points to pad the stats.
Ennis and Kesler were my big misses, given their ice-time and role with their clubs they should have seen more than a “keep pace” rate from the first quarter. I stretched a bit on Purcell, but given that he was only averaging just 15:54 at the time, his production rate probably wasn’t going to skyrocket.
A couple of nice hits in this list. Scheifele, Atkinson, Ryder, Henrique, Santorelli, Read and Barkov made for great waiver wire additions. Bergeron, Sedin, Ribeiro all kept pace with slight increases but nothing to write home about.
Whiffed on a few “big named” guys. Laidlaw is probably right on Heatley being a “corpse”, but I still can’t shake the feeling to write him off completely. Brad Richards took a bit of an offensive hit, but I still consider him a huge 2nd half pickup given his role of responsibility in NY.
Buffalo still perplexes me, how can a player average close to 20 mins a game but score at less than 0.2 point-per-game pace? If they ever figure it out, Ott could be a huge bust out candidate for the second stanza.
The FLA pairing of Fleischmann and Huberdeau also bucked the trend while putting up some unimpressive numbers taking into account the responsibility that they’re being given. They’re shaping up to be an “offense-by-committee” team with equal sharing of ice-time, so be wary of them moving forward.
Most of the players from this list fell into place. A lot of the “hot” starters came back down to Earth in the 2nd quarter. Steen still managed to post 12 points in 14 games before sustaining a concussion. His point production fell to 70 percent of what he produced in the first quarter. If he continues to post a point-per-game rate of 0.86 for the remaining 43 contests, he’ll finish the season with roughly 75 points, which is a nice total in itself, but the major damage is already done. If you’re still hanging onto him, I’d maximize value the best I can and see if you can pawn him off to an unsuspecting owner. Duchene is still chugging along near a point-per-game pace despite averaging just over 18 minutes a game. With Paul Stastny heading into UFA status, the Avs could be open to moving him at the deadline, which will free up the 18:40 to be distributed throughout the line up.
Hossa and the Hawks continue to defy historical numbers, as he’s posting point-per-game numbers despite limited (18:16) overall and PP (2:11) ice-time. His stats are buoyed by the fact that Chicago is averaging a crazy 3.67 goals per game. If the offense dries up, Hossa could be the first to feel the wrath. Consider him a strong “sell high” candidate for the 2nd half. Cogliano could be classified in the same boat. I thought a 0.4 point-per-game pace was elevated for a player of his status, but he’s shocked me by increasing that to 0.78 in the 2nd quarter. The Ducks’ secondary scoring is ridiculous at the moment, but that should taper off as the season progresses.
More on the next page...