There were a few studies that have looked at doing doubles every other day (3 days a week) vs singles every day (six days a week). Volume and "pace" exactly the same.
In the first study I read (this is all from memory) the participants were their own controls, using one leg (on some sort of cycle-like machine, I think more like a constant load "kicking" exercise ) as the "singles" (one workout a day) leg, and the other leg as the "doubles" (two workouts a day, every other day) leg. They did 60 min steady-state workouts once (singles leg) or twice (doubles leg) a day for ten weeks. At the end, the doubles leg had much greater endurance, and had a significantly larger increases in many enzymes and other related blood measurements that are known to increase with increased endurance, than the singles leg, despite the same exact volume and intensity exercise program. The researchers argued that this could be due to the "doubles" leg undertaking more workouts in a glycogen depleted state (second workout, participants did not eat between exercise session on doubles days), which led to a larger stimulus of glycogen re-synthesis (and the doubles leg did have larger resting glycogen stores). However, the extra off days could also have improve recovery .Very interesting study.
A few caveats:
* I think these were fairly untrained participants
* This was a bit of an unusual exercise they were doing
* One leg exercise is not something normally participated in, one always exercise both together for endurance exercise
* The final endurance test was a "test to exhaustion" at a predetermined level of effort (x% of vo2 max or something like that), and these types of test tend to create larger differences than true time trial efforts which are more relevant to the real world.
* most athletes would do doubles or singles every day, not doubles every other day
Despite all that, the results were indeed intriguing.
And later, with experienced cyclists, other researchers retested the same concept but using two groups, one who did 6 singles a week, and one group that did 3 double workout days a week. However, this time, they introduced interval training along with steady state efforts. This pushed up the overall intensity. In this case, a much more relevant, real world scenario, there was no significant difference between endurance of athletes at the end of the protocol. However, doubles athletes did have slightly better endurance, slightly higher glycogen stores, and some higher measurements of other metabolic indicators related to endurance. So the doubles protocol did produce some differences, and most which pointed to a potential endurance advantage, but the actual test to exhaustion did not produce significant results. Yet still interesting that 3 days a week (3 doubles that is) was just as good, if not better in ways, than a standard 6 singles. I am sure they have done other spins on this concept (in fact I know they have), but I need to look up results.
Lastly, I once saw a re-print on line of some Lasse Viren training: lots of doubles and even triples, which included many moderate-distance runs of 3-5 miles.