Learning from Rating mismatches

How should someone think about improving when their ratings in different types of chess are vastly different? For example, my Daily chess rating is more than 300 points higher than my Live chess rating, even when I'm playing 15|10 time control (so it's not just that I'm bad at bullet). It's not that the scales are different, as I've checked my relative percentiles across Chess.com in each.

How should I interpret this if I want to get better at live chess? Does this discrepancy suggest that I'm bad at time management or something? Maybe that I rely too much on looking at concrete variations on the analysis board? If this data suggests that I might beat the same person in daily chess that I would lose to in live chess, how do I use that knowledge to play better live chess? Similar questions could be asked for discrepancies between live and tactics ratings.

I'm curious how many other people are in similar situations and how they interpret this.