Once a sim has good puck and player physics, and good AI in all categories, is the hockey game engine complete except for the details? No. There's one other subject I'd like to discuss. Momentum. This is something that has been ignored by developers of sims (for all sports) for too long. Everyone who plays or watches sports is familiar with how the flow of the game shifts back and forth, with one team or the other having the momentum at any given point. When your team scores a goal (or gets a big hit, pulls the goalie, kills a penalty, wins a fight, etc.), it lifts the spirits of the entire team and they "turn it up a notch." This might seem like a difficult thing to implement in a sports game but in fact it is not.
One of the features which gave WGH an almost uncanny sense of realism was the implementation of momentum. This was implemented by having a "momentum factor" for each team, which affected the speed attribute of all players on the team. The momentum factor was cumulative, so that if you scored three goals in quick succession the other team would seem to be skating in mud. The momentum advantage wore out gradually (i.e., the momentum from scoring a goal might last two minutes) so you had to take advantage of it quickly, just like in real life. If the losing team did manage a goal then they would get the momentum boost, giving them a chance to get back into the game. Momentum is mutually exclusive, only one team can have it at a time - so any amount added to one team's momentum factor is subtracted from the other's. As well as giving the game a natural ebb and flow, this feature also forced you to adopt realistic strategies, such as playing defensively when you have just been scored on. I'd love to see a modern hockey sim adopt this brilliant feature.
The Hockey Wish-List
· Customizable camera angles and zoom level. Two recent hockey games (Powerplay 98 and Actua's Ice Hockey 2) have stumbled badly by failing to include an overhead view that was zoomed out far enough to see the play developing. Rather than risk annoying users with a limited set of views why not allow the camera views to be customized? They've gone to all the trouble to implement a 3D graphics engine, they should let us take advantage of it.
· A career mode is another common request. People want to develop players and teams over multiple seasons, including the use of entry drafts, trades, retirements, career statistics, and injuries. Ideally there would be a financial model including team-specific spending caps (i.e., not all NHL teams have the same amount of income) and contract negotiations. Also, it would be cool if you could collect Stanley Cup banners in your rink and retire the sweaters of star players.
The trading should be intelligent, not a simple comparison of how many skill-points you are offering. For example, if you offer a good goalie to a team than is getting shellacked every night, they should be willing to give up more for him than would a team with Patrick Roy. Computer teams should trade amongst themselves. And even when trading via rating-comparisons, keep in mind that in real life GMs can't see exact numerical ratings of players capabilities. In fact they look more at a player's stats than anything. So if I develop a poor player into a forty-goal scorer, I should be able to get someone good for him regardless of his numerical attribute ratings.
· Realistic simulated games. Even with realism in the games you play, the stats will not make much sense if all the simulated games in a season are 1-0 scores.
· A nice feature in Powerplay 98 was a "playoff performer" player attribute. Some players are better in the playoffs.
· As previously mentioned, there should be an arcade/simulation option, or better yet a slider bar that lets the user specify the speed of the game.
· Different reaction to hits, such as getting up fast, getting up slow, going directly for a line change, neither player falling (i.e., they just lose speed).
· When players fall or go down to block a shot, make them slide.
· A player that causes an injury should get a major penalty or even a suspension.
· A play editor for setting up custom plays and strategies.
· Real hockey strategies, such as forechecking with three players and cycling the puck around the boards until there is an opening in front of the net.
· Realistic coaching strategies that affect the on-ice play.
· More crowd involvement, especially audio involvement: having the crowd cheer or boo at appropriate times, or counting down the last few seconds on the clock if the home team is winning in a playoff game.
· 3 parts Powerplay 98 (the teammate AI, the coaching options, the gamepad control/skating model)
· 3 parts Actua Ice Hockey 2 (the offensive computer AI, puck physics, and goalies)
· 5 parts Wayne Gretzky Hockey (the momentum implementation, player physics, game speed, defensive computer AI, and general hockey smarts)
One final word of advice for hockey game developers. There is a classic test for determining the complexity, authenticity, and quality of a sports simulation. You take a novice, someone with little experience in the sport, and see how long it takes them to master the game (for a serious sim, it will probably take several months of intensive game time). Then you take an expert in the sport, someone who's competed at a high level, and sit them in front of the game. Including learning the physical controls, they should be able to compete very well in the game after just a few hours. The developers of Wayne Gretzky Hockey created a sports simulation that passed this test. Can you?