Are you asking if they're on all the state ballots? Obviously not, but neither is Jill Stein.

I'm not trying to argue here. Obviously there will be a lot of people who will throw their votes away on Johnson or Stein, particularly around here, so it makes perfect sense to discuss them.

I'm just trying to correct this technical point, about how polls, supposedly by definition, include all the candidates. Now you're saying most reputable polling organizations include Stein and Johnson. More accurately: most reputable organizations run the polls both ways, because both kinds of polls give different kinds of information. Polls that include the major third-party candidates are useful for judging their grass-roots support, and in particular whether it's worth including them in debates. (It gets tricky even then to judge the cutoff for the polls, though. Should they include the Socialist Workers Party? The Constitution Party?) But polls that don't include third parties are generally more accurate for estimating the final election outcome, at least historically.

There is some difference of opinion about which kind of poll is better for forecasting. Fivethirtyeight mostly uses polls that include Johnson and Stein and simply tries to correct for the third-party effects in their statistical models. Upshot, Princeton, and the others stick with head-to-head polls. This is sometimes cited as a reason fivethirtyeight generally gives Trump better odds than the other aggregators.