(Ed Mulholland-USA TODAY Sports)

The New York Rangers have life. Goaltender Henrik Lundqvist kept the Blueshirts from being swept out of the Stanley Cup finals with a 40-save performance in a 2-1 victory against the Los Angeles Kings in Game 4, but the biggest save of the night would come from the soft ice in Lundqvist’s crease.

“Obviously, I don’t want it to go in the net, so I’m just trying to do whatever I can to stop it,” Rangers forward Derek Stepan said. “After I pushed it back under [Lundqvist], I just don’t know where it was going or what was going to happen. It was kind of a lucky play.”

“Thank God for soft ice now and then,” Rangers coach Alain Vigneault joked after the game.

Winning the rest of the way is still a tall task for New York: according to WhoWins, of the 21 previous higher-seeded teams to take a 3-1 series lead in the Stanley Cup finals, all have won the Stanley Cup. Those teams have a 13-8 record in Game 5 as well.

But tough isn’t impossible.

Using a teams goals scored and allowed we can make a best-guess estimate of what a team’s winning percentage should be. The Kings, for example, score 3.4 goals per game in the playoffs and allow 2.7. By using the Pythagorean Winning Percentage formula we can estimate their winning percentage to be .609, or for them to win roughly 61 percent of their games. The Rangers should win 54 percent based on their goals for/against ratio. In the 1981 Bill James Baseball Abstract, James introduced the log5 method to answer the question, “how often should team A be expected to beat team B?” I will spare you the gruesome math behind it, but here is the formula.


Using this formula, if you have a .610 team playing a .540 team, this method shows that the better team can be expected to win 57.1 percent of the games between these two teams.

So what are the chances New York wins the next three against Los Angeles? If the Rangers have a 43 percent chance of winning any particular game then the probability they win three in a row is .43 times .43 times .43, or 8 percent.

But what if one of the teams gets “hot?” How does that effect the Rangers chances of a comeback?

I decided to use a Monte Carlo simulation, which is a technique used to approximate the probability of certain outcomes by running multiple trial runs – or simulations – using random variables. I built in to the model the chance that a team could get hot for any particular game. You wouldn’t know in advance when or if the team would go on a heater, but the opportunity is there. Same for going “cold.” I assigned the Rangers with an arbitrary chance at going hot or cold at 20 percent, which increased their chances of winning by that amount as well. I could just as easily used 30 percent or 10 percent for the change of state so don’t read too deep into those variables. I then ran the simulation 10,000 times and counted the number of simulated series where the Rangers won the next three games. Here are the results:

So yes, I am saying there is a chance. Just not a very good one.