THN.com/podcast. From THN On The 'A': Generating AHL Analytics and Rating Prospects with Sean O'Brien Connor Bedard [https://thehockeynews.com/.image/c_fit,h_600,w_600/MjAwNDcxNjAxMTgxODkzNzQw/usatsi_20963261_168400847_lowres.jpg] Connor Bedard A year ago, I defended a thesis that centered on finding inefficiencies in the NHL draft, determined strictly by mathematical means. While there is always context to be provided, the goal was to build a foundation of academic research so that I could build a few models that measure draft success, development success and predict future careers of various players. There is much to be learned, but I have built a model that gives me the best-drafted prospects with fewer than 25 NHL games played. Some teams will have multiple players, and some teams won’t have any. Every model is different, and I always employ the cliche of my first statistics professor, “All models are wrong. Some are useful.” That is to say, no one’s model is definitively correct, and there is always room to improve. Different models will value different things. I’ve worked with a few NHL and MLB teams on various prospect projects, and each team has a different model that values different outcomes. The goal is to remove scouting bias and preconceived notions about players to gain a better understanding of what they project to become based on a variety of factors. This model, named W5M, ranks each NHL-affiliated prospect based on their expected career output. It should be noted that goalies are not included in this ranking, though they are included in the model. Some players will be bright flashes in the pan, while some will have 15-year careers as a sixth defenseman. Regardless, this model favors star potential, meaning some players with higher ceilings and lower floors will rank higher than those with high floors and low ceilings. While the high-ceiling players may not be as sure of a thing, their potential is weighed against the likelihood of achieving it to appropriately rank them. The model accounts for production, the production curve, league played in draft year, league played in currently, strength of team, volatility, games played (injury), height and weight. The model does not account for time on ice, as it cannot be verified as accurate across leagues. It does not consider international tournament performance, such as the U-18s or world juniors, as more important. However, the strength of the team does change, and the production is weighted accordingly. No adjustments were made to the rankings based on scouting, and while I would personally change the order of a few players, the point of the model is to gain a general understanding of a player’s most likely career output. Without further ado, here are the top 50 drafted NHL prospects based on this model. 1. Connor Bedard, Chicago, Top Line 2. Adam Fantilli, Columbus, Top Line 3. Matvei Michkov, Philadelphia, Top Line 4. Leo Carlsson, Anaheim, Top Line 5. Logan Cooley, Arizona, Top Line 6. David Jiricek, Columbus, Top Pair 7. Zach Benson, Buffalo, Top Line 8. Luke Hughes, New Jersey, Top Pair 9. Pavel Mintyukov, Anaheim, Top Pair 10. Ryan Leonard, Washington, Top-Six Forward-plus 11. Simon Edvinsson, Detroit, Top-Four Defenseman-plus 12. Brandt Clarke, Los Angeles, Top-Four Defenseman-plus 13. Will Smith, San Jose, Top-Six Forward-plus 14. Shane Wright, Seattle, Top-Six Forward-plus 15. Olen Zellweger, Anaheim, Top-Four Defenseman-plus 16. Simon Nemec, New Jersey, Top-Four Defenseman-plus 17. Marco Rossi, Minnesota, Top-Six Forward 18. Dmitriy Simashev, Arizona, Top-Four Defenseman 19. Oliver Moore, Chicago, Top-Six Forward 20. Cutter Gauthier, Philadelphia, Top-Six Forward 21. Dalibor Dvorsky, St. Louis, Top-Six Forward 22. Lane Hutson, Montreal, Top-Four Defenseman 23. Matthew Knies, Toronto, Top-Six Forward 24. David Reinbacher, Montreal, Top-Four Defenseman 25. Frank Nazar, Chicago, Top-Six Forward 26. Logan Stankoven, Dallas, Top-Six Forward 27. Denton Mateychuk, Columbus, Top-Four Defenseman 28. Jimmy Snuggerud, St. Louis, Top-Six Forward 29. Marco Kasper, Detroit, Top-Six Forward 30. Owen Pickering, Pittsburgh, Top-Four Defenseman 31. Jiri Kulich, Buffalo, Top-Six Forward 32. Mikhail Gulyayev, Colorado, Top-Four Defenseman 33. Andrew Cristall, Washington, Impact Forward-plus 34. Calum Ritchie, Colorado, Impact Forward-plus 35. Alexander Nikishin, Carolina, Top-Four Defenseman 36. Quentin Musty, San Jose, Impact Forward-plus 37. Matt Coronato, Calgary, Impact Forward-plus 38. Kevin Korchinski, Chicago, Top-Four Defenseman 39. Matthew Wood, Nashville, Impact Forward-plus 40. Axel Sandin Pellikka, Detroit, Top-Four Defenseman 41. Gabe Perreault, New York Rangers, Impact Forward-plus 42. Joakim Kemmell, Nashville, Impact Forward 43. Matt Savoie, Buffalo, Impact Forward 44. Brock Faber, Minnesota, Top-Four Defenseman 45. Brennan Othmann, New York Rangers, Impact Forward 46. Tyson Foerster, Philadelphia, Impact Forward 47. Noah Ostlund, Buffalo, Impact Forward 48. Danila Yurov, Minnesota, Impact Forward 49. Stanislav Svozil, Columbus, Top-Four Defenseman 50. Liam Ohgren, Minnesota, Middle-Six Forward
