The race at this point is stable and some outliers have probably rolled off. Biden lost about half a point in the 538 average as some people went from wanting to vote for him to probably not voting if the election was already here.
When one is behind in a competition of any kind, then stasis ensures defeat. A stable 8% lead for Biden nationwide implies a win of about the same scale as Obama in 2008 in which Biden has much the same electoral map as Obama in 2008 except for trading Indiana for Arizona or Georgia. (Picking up both puts him at the level of Bill Clinton in 1996, but with a very different map from Bill Clinton).
A telling fact: Biden is consistently about 10% behind Trump in Indiana (or perhaps a bit less - high single digits). Republican nominees for President have not won nationwide without winning Indiana by at least 10% for a century. The two times in which they won the electoral college without winning the popular vote, 2000 and 2016, were times in which they won Indiana by huge margins. Maybe the reasons have changed, or maybe it is that Ohio is typically a deciding state, and Ohio is never quite as R as Indiana.
How valuable is stasis? It is worth a strategy that surrenders a lead slowly but not enough so that the calendar runs out. I think of Tom Landry's nickle-back defense when his Cowboys had a run of twenty-some consecutive winning seasons. The Cowboys might get a quick lead of 20 points, but after that, Landry used the clock as an ally. To undo an early 20-point deficit an opposing team must score quickly, which means that the running game is inadequate for that. On offense Landry went to the five-yards-and-a-cloud-of-dust running game, which is good for eating the clock and perhaps getting a field goal that adds a bit to the lead. It also kills about ten minutes. On defense, Landry had the nickelback defense... five defensive backs in the backfield that killed the opposing passing game. It made any but short passes ineffective, and short passes might get five or six yards. An offense needs two such plays to get a first down, and it needs more than that to get the quick scores.
So the team behind tries a high-risk pass play to get one of those game-changing big gains that gets twenty yards or so. There's a Cowboy back to intercept it. Should there be a score on that play it is highly possible that it is a touchdown by one of the Cowboy's defensive backs. Things get far worse for the team that was behind by 20 points.
