Quoting: buds16
You replied with advanced stats as proof? first pdo is luck, so your reply has no credibility. Real stats make your reply even less relevant.
Polak. 54 GP, 2G, 10A +5 46PIMS 0.22PPG 1 MILLION
Gudas . 70GP, 2G, 14A, 0+/- 83PIMS 0.23PPG 3,350,000 MILLION
WOW.
PDO is context - Gudas had worse puck luck than Polak while posting better numbers. Points per game, plus minus, or penalty minutes don't really have any impact on who is better. Advanced metrics are "real stats" - in fact, they are more valuable for individual player evaluation than points, +/-, and penalty minutes.
When you delve into the underlying numbers, it's abundantly clear that Gudas makes his pair-mates better, while Polak makes them worse.
Looking at the
Gudas-Sanheim pair (238 5v5 mins), they posted a 7.84 rel CF%, 7.81 rel xGF%, along with an abysmal 94.9 PDO (atrocious puck luck)
But when you examine
Sanheim-MacDonald pair (208 5v5 mins), they posted 5.35 rel CF%, 6.04 rel xGF%, but had an upper tranche ranking in PDO of 103.3
Moving over to Polak, him with Dermott (223 5v5 mins), they had a 3.07 rel CF%, -1.14 rel xGF%, and a 103 PDO.
However, when you look at Dermott with Carrick (184 5v5 mins), the numbers are vastly better : 7.4 rel CF%, 14.72 rel xGF%, and PDO of 96.
Looking at Owen Kewell's NHL Similarity Tool, Gudas' most comparable player for offense is Calvin de Haan (70%) and Stephen Johns for defense (87.7%). Roman Polak's most comparable player for offense is Kevan Miller (89%) and Luke Schenn (77.4%) for defense.
If you don't believe in the value of underlying metrics, that's your prerogative. In doing so, you will loose respect to have a rational conversation if you refuse to acknowledge that they are realistic metrics to determining individual player analysis.