Written by GarikTuesday night, Jack Capuano made perhaps his dumbest decision ever. I'm not even going to try and use any nicer words about it, the decision hurt the team's chances of winning that game in Toronto and didn't do anything to improve the chances of winning in the future. By definition, that is the worst type of mistake a coach can make.I'm talking of course, about the scratching of Michael Grabner.Grabner was scratched, ostensibly because he failed to produce any offensive numbers (points) in each of his last 12 Islander games. Capuano thought that giving him a night off might somehow encourage him to score more.The problem with this, first and foremost, is that Grabner is a lot more than a scorer. He's also one of the Islanders best defensive forwards.With Grabner on the ice, the Isles have faced 24.5 shots against per 60 minutes, second best amongst Islander forwards (Josh Bailey is No. 1). If you use unblocked shot attempts instead (shots on goal and missed shots, also known as fenwick), Grabner leads the team. The Isles have faced only 33.7 shots against per 60 with Grabs on the ice.The end result of this is that the Islanders have outshot opponents 115-99 with Grabner on the ice at 5-on-5, the best performance of any Islander. In short, with Grabner on the ice, the puck is more frequently heading into the offensive zone and away from the Islanders own net. Even if he's not scoring, he's doing a hell of a job preventing the opponents from scoring. Islander goalies have an .879 SV% with Grabner on the ice at 5 on 5 (HORRIBLE), and Grabner is only -1 at 5-on-5. That's impressive.Oh, and that's not a change from previous years either.Moreover, Grabner is an ELITE penalty killer, by far the best the Islanders have. With Grabner on the ice, the Isles have allowed 8.3 goals against per 60. That's over 2 goals better than the next best Islander forward (Matt Martin at 10.5 goals per 60).If you look at shots, it's even more extreme. With Grabner on the ice at 4-on-5, the Isles are allowing only 37 shots against per 60 minutes. The No. 2 penalty killer -- Frans Nielsen -- is allowing 48 shots per 60, 11 more.In short, by removing Grabner from the ice, you're forcing the team to play without one of its best players at keeping the puck from heading into the Isles' own net, something that's pretty important given their goalie situation.But wait, not only is Grabner one of the Isles' best defenders, but he's also one of the team's best scorers as well -- even after tallying no points in 12 games.Did you know that Grabner has been the 3rd most efficient even strength goal scorer over the past three years in the entire NHL? Well, he is. He ranks behind only Steven Stamkos and Jonathan Toews in goals per 60 minutes at 5-on-5 (and ahead of guys like Rick Nash or Evgeni Malkin or every single Islander). Three seasons worth of data is a better indication of Grabner's scoring potential than the results of the last 12 games.Moreover, even looking at just this season, did you know Grabner is second among all Islanders in points per 60 at EV? Only Nielsen is putting up points at a better rate. But how is this possible, you ask? After all, many other guys have been putting up points these past 12 games.Sure, but these other guys are managing to put up points on the power play. Grabner is TWELFTH in power play time on ice per game this year, behind such marvelous goal scorers as Colin McDonald, Peter Regin, and Cal Clutterbuck. You can make a case Grabner doesn't deserve much PP time, but understand that doing that is going to make it a hell of a lot more likely he puts up less points and goals every now and then, because those are the best opportunities to get points.In short, Grabner is not only a great defensive forward but odds are he's still one of the team's best scorers, and benching him because he has not scored in 12 games only makes our chances of scoring in the next game WORSE, not better.In conclusion, I must ask, what is the purpose of benching a player? There are a few reasons to do so.The first reason is simple and unquestionable: in favor of a better player. This is what happened when the Islanders were scratching Colin McDonald in favor of Brock Nelson and there's no problem with that. But Colin McDonald and Eric Boulton are playing over Grabner. There is no conceivable way you can make the case that either McDonald or Boulton is a better player on offense or defense than Michael Grabner. There is no question: by scratching Grabner, they made the team worse Tuesday night.The second reason is a little more questionable: To teach a lesson to a player who needs to learn something.In this case, you're making your team worse to make the team better later on. In some cases, this can be reasonable! If a player repeatedly makes bad changes or decisions in a game -- especially after he's been told not to do so -- scratching him the next game to send a message can make sense.But what exactly was the message being sent here? Shoot better, Grabner? Make Regin and Clutterbuck into better offensive threats? It's hard to see how these are things that can be fixed by taking a game off. A slump is more likely fixed by getting out there and continuing to create scoring chances (Grabner created a few in the previous game vs Detroit, even if he didn't convert). Eventually a few of those will go in. Taking him out when his replacements won't even create those chances is a dumb idea.In short, the Islanders removed one of their best defensive players (who needs defense?) and one of their better scorers for players who are worse scorers (McDonald has less points than Grabner despite Grabner's slump!) and worse at defense for no reason.That's really dumb.