This is the time of year you'll see various prognosticators run millions of simulated seasons to come up with their baseball predictions. I'm generally not a big fan of ultra-extended simulations, because 1) you're not going to get all that much better results by running a million simulations than you will with a thousand or even a hundred, and 2) simulated results are only as good as their inputs, anyway. What good are the projections you came up with for the 2010 Mets if every one of the 1.2 million simulated seasons had Jose Reyes playing 160 games?

But as a year goes on, and a simulation's possible results telescope into a narrower range, it becomes a lot more useful. Right now, there are about 15 games left in the NHL season, few enough that it's reasonable and telling to forecast all the remaining results and see what we come up with. And I wanted to show you two sites doing just that.