Other things to note about this combo:
can become a one sided board wipe if something like a Howling Mine
or Font of Mythos
is in play - you can discard your hand without regret.
-- Avatar of Discord
's discard requirement is much less harmful under the same conditions
ends the game. I played another 2hg game yesterday and my random partner had a poorly built mill deck, but he did have Traumatize and he was willing to wait until turn 6 to cast it for the 3rd Bloodchief counter to activate on turn 5. Our opponents were around 25 life or so, didn't see it coming...they hadn't even been milled very much until then. On the other hand, Traumatize probably isn't necessary if you're running turbo-mill because every other mill card you have probably makes a win inevitable with Bloodchief. You don't have to deck them to win - you realistically just have to mill 10 cards once Bloodchief is active, unless you're playing a lifegain deck.
--even low-grade mill cards can turn the game around if Bloodchief is active. Cards like Cephalid Broker
mean 4 damage and 4 life a turn, every turn
--Jace's ability to bounce / crippling chill / sleep blockers makes it fairly easy to get damage through to activate counters
--discard isn't a very viable strategy for GM if your partner is playing with Howling Mines so GM probably should be adjusted to account for that.
I don't know how good this is overall because it's reliant on a card for which there are only 3 copies, and it takes 3 turns to activate it so enchantment removal can ruin your day. On the other hand, even without the Bloodchiefs, Jace seems like a decent overall combo (providing card draw for a deck that greatly benefits from it). A turbo-mill/control hybrid should mean Jace can remove blockers for Rakdos, provide card draw, and turn Bloodchief into an almost instant win condition.
I've only played this with random partners; I don't have a regular partner and I haven't created any specific builds - I think the total number of times I've played this combo is 4, which isn't a lot of data. But those 4 times were pretty easy wins, except for the first one, where we were behind and stumbled onto this by accident, but won shortly thereafter.