There's got to be some mathematical formula that can take how easy it is to get points and convert that to how easy it is to have a scoring streak, but I can't think of what it would be. It wouldn't just be proportionate. If it's twice as easy to get points in that era, it wouldn't quite be twice as easy to get a scoring streak I imagine, it would be a little less.
If we are willing to make some simplifying assumptions it's possible. The crucial assumption is independence. That is, the probability of scoring a point in a given game is independent of scoring another point in that game. Also there is no correlation of point scoring between games. This assumption is unlikely to be true (strength of opposition differs, injuries to teammates and, potentially, hot streks) but we may be willing to make this simplification. If so, the probability (following the poisson distribution) of being held without a point in a game is e^(-a) where a is the average number of points a player scores in a game.
For instance, scoring at a 120 point pace over 80 games, that is 1.5 ppg on average, the probability of being held scoreless is 0.22. That means the probability of getting at least one point is 1-0.22=0.78. The probability of getting at least one point in 25 straight games is then 0.78^25=0.00181, around 0.181%.
Now, we want to compare this with scoring in 51 straight games (Gretzky's record) but during a higher scoring era. So far this season, the average team scores 2.76 goals per game. During 83/84 the average team scored 3.94 goals per game. Making the simplest possible adjustment, each point today would be worth 3.94/2.76=1.43 points back then. 120 points today would then translate into 171 points.
(This may be an overestimation of how hard it is to score today, see for instance this thread:
http://hfboards.com/showthread.php?t=801447)
Scoring at a 171 point pace (over 80 games) eqauls a ppg of 2.15. The probability of being held scoreless a given game is then e^-2.15=0.12 which means the probability of getting at least one point is 1-0.12=0.88. So, the probability of getting at least one point for 51 straight games at that pace is 0.88^51=0.00175, or around 0.175%.
So, in this example, it is about as likely to score in 51 straight games in 83/84 as it is to score in 25 straight games this season. However, the probabilities depend crucially on which scoring pace we assume. For instance, scoring at a 2ppg pace this season gives a probability for a 25 game scoring streak of 2.64%. That translates into a ppg pace of 2.86 in 83/84 which gives a probability for a 51 game scoring streak of 4.96%. The following graph shows the probability of each streak respectively where the ppg pace for this season is shown on the x-axis.
As can be seen, as the scoring pace increase, the probability of the 51-game streak in 83/84 increases faster than the probability of the 25-game streak this season. I've got to say, I'm surprised by the result of this calculation. I expected Gretzky's streak to be harder than Crosby's. On the other hand there are a lot of assumptions (independence, era adjustments) that may not be ok so it should be taken with a grain of salt.