Tony Vargas
Legend
Nod, I'd just as soon narrate success, at that point, or cut to the chase and have the roll determine not success/failure, but time (or some other resource) to accomplish.For any task with an uncertain outcome and a meaningful chance of failure where one could try the same approach to a goal repeatedly, the player can trade time in the amount of 10x the normal amount of time it takes to complete the task for automatic success.
Mostly a very on-the-money article, though I disagree a bit about the topic of group checks.To put an abrupt end to the "piling on" effect [MENTION=996]Tony Vargas[/MENTION] describes above, we've been employing Angry's rules for Teamwork and Group Checks (http://theangrygm.com/tweaking-the-core-of-dd-5e/). It has worked out quite well in multiple campaigns.
One thing I think was missed that the multiple d20 rolls of a repeated check that only needs one success to win or, very alternately, one failure to blow, is that, while they may make superficial sense, narratively, statistically they add up to near-certain success or failure, respectively, and not because of the best or worst character in the party (that becomes almost meaningless) as 'weakest link,' but just because of the sheer number of rolls - it's an obvious/intuitive bit of modeling that fails as a model of uncertainty.
That's why Group Checks actually make a fair bit of sense: they give the party a reasonable chance of success at a group task, whether that's reaching a consensus on a knowledge check (rather than just trusting the 'expert' with the best check), or reaching the other side of the cavern without waking a dragon.