Goalies hate defending it and Winnipeg Jets defenseman Dustin Byfuglien thinks it's terrible, but so far the NHL's experiment with 3-on-3 overtime is doing exactly what it was intended to do -- significantly cut down on the number of games that have to be decided by a shootout.

It's still early, but through the first three weeks of the season the NHL has pretty much cut the number of games that have required a shootout in half from last season. As of Wednesday, 27 NHL games have required overtime this season with only nine of them (just over 33 percent) needing a shootout to determine a winner.

Compare that to the same point last season when 21 of the 35 overtime games (60 percent) remained tied after five-minutes of 4-on-4 hockey. That is a pretty significant drop, and it has to be pleasing for general managers that were tired of having so many games be decided by a shootout instead of something that resembled an actual hockey game.

Before this change the NHL had already taken steps to reduce the importance of the shootout by trying to convince teams to take more chances in overtime by removing shootout wins from the tiebreaking procedure for playoff spots. By taking four players off of the ice (two on each team) and opening up the game, the shootout has been dealt another pretty significant blow, while also adding some excitement to a league that has seen its skill and creativity get shut down by structured defensive play and all-time great goaltending. 

What else have we learned from our first look at 3-on-3 hockey? Let's take a quick look.

Goalies are performing poorly, but it hasn't had a huge impact on their numbers ... yet

As expected, goalies are having the most difficult time adjusting to the overtime change and have managed only an .846 save percentage during 3-on-3 hockey.

That is obviously the result of having to face so many breakaways, odd-man rushes and scoring chances with all of the extra ice that is available to the skill players in front of them. In anticipation of that, New Jersey Devils goalie Cory Schneider suggested before the season that the league should keep 3-on-3 stats for goalies separate from the rest of their stats because of the impact it could potentially have on them.

But here's the thing: So far it hasn't made that much of an impact on their overall numbers.

The league average save percentage is still .915, the exact same mark it was a year ago (which was the highest it's ever been since the stat has been kept by the league), while the NHL goals per game average is down (again) from where it was a year ago.

And that really shouldn't be a huge surprise, should it? Only a small percentage of a team's games are going to require overtime, and they're not all going to end with their goalie giving up a goal. There's also the fact that several games have ended on power play opportunities (which become 4-on-3 situations) and would not be impacted by 3-on-3 totals.

The individual goalies that have been impacted the most so far are Ryan Miller of the Vancouver Canucks and Steve Mason of the Philadelphia Flyers.

Miller has been on the losing end of three overtime games, stopping just two of the five shots he's faced during 3-on-3 play. If you remove his overtime stats his save percentage for the season goes from .930 to .935.

Mason has lost two games (stopping 9 out of 11 3-on-3 shots) and would see his save percentage go from .899 to .905 if the numbers were kept separate.

Other than that, nobody has really been impacted (yet) when it comes to their personal numbers and at the end of the year the league is still going to see huge goalie numbers and low-scoring games. 

Protecting the puck is essential, and your shots better hit the net

With so much open ice and teams only using their most skilled players, avoiding turnovers is a critical part of winning in these situations.

Also important: Making sure your shot attempts hit the net, because anything that doesn't pretty much serves as a turnover and almost instantly starts an odd-man rush the other way.

So far six of the 3-on-3 overtime goals this season have been scored within eight seconds of a either a turnover/giveaway, or a missed or blocked shot at the other end of the ice. Another missed shot resulted in a penalty shot going the other way, and that's not even taking into account the odd-man rushes and chances that didn't result in a goal.

Even if teams try to take a "safe" approach to these situations there is so much open space on the ice that it's not going to take much to create offense. And once that offense starts happening and teams start exchanging chances it's simply a matter of when, and not if, a goal is going to get scored.

The average game-winning goal in overtime this season has come within the first two minutes of the overtime period, including five games that have ended within the first minute.

The other noticeable development so far has been the number of penalties that have been called in overtime because, well, desperate times call for desperate measures.

With so much space for players to make plays with the puck and it being so easy (relatively speaking) for players to blow past defenders, it's resulted in them doing whatever they can to slow their opponents down, whether it be a hook, a trip, or just flat out grabbing on to them and holding them. So far nine of the 27 overtimes games this season have had at least one penalty or penalty shot called (33 percent), with four games ending with a four-on-three power play goal. Only 27 percent of the NHL's overtime games at the same point last season had a penalty or penalty shot called in overtime.

Ondrej Palat celebrates an overtime goal for the Tampa Bay Lightning. (USATSI)