Years of conversation fill a ton of digital pages, and we've kept all of it accessible to browse or copy over. Whether you're looking for reveal articles for older champions, or the first time that Rammus rolled into an "OK" thread, or anything in between, you can find it here. When you're finished, check out the boards to join in the latest League of Legends discussions.
Comment below rating threshold, click here to show it.
If you have runes, masteries, and ninja tabi your dodge chance is about 21%. Lets say it takes 10 dodge-able attacks to kill you. In ten attacks, with 21% dodge, 55% of the time you will dodge 1-2 attacks and you will dodge 3-6 attacks about 35% of the time. For those who can't add, that means you won't dodge less than ten percent of the time.
If you don't want ninja tabi, you will only have about 9% dodge. This increases your chances of not dodging in ten attacks to almost 40%. Another 40% of the time you will dodge once. Seventeen percent of the time you will dodge twice, and you will only dodge more than twice about 3% of fights (assuming ten attacks per fight- remember).
Ninja tabi alone offers 12% dodge chance. This bonus confers a 61% chance to dodge 1-2 out of ten, 28% to dodge zero, and 11% for 3 or more dodges.
What does this all mean? It means that dodge runes are going to make a large impact a small amount of the time, and if you get ninja tabi it stands a good chance of saving you outright. It is questionable whether dodge is really worth it if you don't get the boots to go along with the runes and masteries. 80% of the time there will be 0-1 dodges. Not so hot. Of course the other way to look at it is by purchasing only ninja tabi you might survive an additional 10% of fights with crazy dodging (but this is hardly good strategy to win consistently).
My advice would be to stack dodge if you take the runes. Stacking dodge will frustrate your opponent about 1 out of 3 fights. Otherwise you're better off taking HP or armor runes.
For those curious, I calculated these numbers using the Binomial Distribution.