Nov 19 2008
Exploring the Even-Odd rule
Warning! Math ahead! I have taken 2 stats courses in my life, both at intro level. Sorry for any mistakes I have made.
I mentioned the Star Trek even-odd rule before, and I actually had this arguement with a friend after he read my post. Basically the rule states the odd movies suck, even ones rule. I say it’s true, he says it isn’t. Lets look at some of the most recent numbers:
| IMDB | Rottom Tomato (Freshness) | Rotten Tomato (Rating) |
| 2: Wrath of Khan, 7.7 | 2: Wrath of Khan, 92% | 2: Wrath of Khan, 7.9 |
| 8: First Contact, 7.6 | 8: First Contact,91% | 8: First Contact, 7.3 |
| 4: The Voyage Home, 7.3 | 4: The Voyage Home, 86% | 4: The Voyage Home, 6.9 |
| 6: The Undiscovered Country, 7.2 | 6: The Undiscovered Country, 84% | 6: The Undiscovered Country, 6.8 |
| 3: The Search for Spock, 6.5 | 3: The Search for Spock, 77% | 3: The Search for Spock, 6.3 |
| 7: Generations, 6.4 | 9: Insurrection, 56% | 1: The Motion Picture, 5.9 |
| 9: Insurrection,6.4 | 1: The Motion Picture, 54% | 9: Insurrection, 5.8 |
| 10: Nemesis, 6.4 | 7: Generations, 51% | 7: Generations, 5.6 |
| 1: The Motion Picture, 6.2 | 10: Nemesis, 36% | 10: Nemesis, 5.2 |
| 5: The Final Frontier, 4.8 | 5: The Final Frontier, 18% | 5: The Final Frontier, 3.9 |
Nemesis is the only real out of placer in all of these, and for the top and bottom, the different rankings seem to agree.
10 is a small sample, even using 3 sources (so, a sample size of 30), so does this outlyer ruin our theory? I took the average of the ranking order, and ran some statistical tests on them.
First up, the t-test, with p=0.05
I got a t-score of 0.004574, and to pass it must be greater than 1.7. So, a fail here.
Next I ran a f-test, again using a critical value of 0.05
The result was 0.7852, and needed greater than 2.4 to pass. Not so much.
The results are pretty clear, the outliner of Nemesis destroyed the rule statistically. Unfortunately, when I ran the same tests excluding Nemesis I still got a fail… guess we don’t have enough data.
Ignoring statistics and looking at the trends, it’s pretty clear that the rule is fairly firm… in the past. Nemesis was a steaming pile of crap, only outdone by The Voyage home.
Since we’re ignoring stats, lets have a graph!
Here we go, now we see the rule. Up, down, up, down, etc, until we get to the end where we have a little slide.
Anyway, what does this all mean for the upcoming movie? The numbers don’t really tell us honestly. The rule could have been broke, or reset, with 10. Maybe Trek operates on a base 9 system? Maybe 10 never really happened?
I still fear for this movie, but I don’t trust the even-odd rule anymore.
