I am not really contradicting you here. Just want clarification from GP. My reaction was Huh? As well.

With N=15 years of gps data and 40-70 of those days per year below zero very few below -10°C. I can say that I ride slower on cold days on avarage.

I am not fit and do not track that at all. I can barely keep my avarage above 25km/h in city traffic for 15km, very few conflict points maybe twenty but not comparable to road cycling. So a km/h decrease of avarage speed during winter is common but not a given. I do not count bad road conditions, but Stockholm has really good maintanance during winter, ice free asphalt all year.

More wind chill? More weight by thicker clothing?

I ride with heavy loads, 10+ kilos is common, no stats on that though. For my speed it does not matter much with 2 kg clothes.

I definitely ride slower in winter too but I always assumed that was due to my body having a harder time in the cold rather than air density changes.

That seems to be born out by using better (read more expensive) winter riding gear. I naturally run a bit cold so spending on low temperature tights, base layer and jersey/jacket has made a massive difference for me.

Track your time on a consistent route on paved, snow free roads. I guarantee you'll see a dip in winter.

Scorching heat with high humidity has an opposite effect. H2O drops the density of air and you can go measurably faster than in drier air.

Interesting! I’m also a long time high volume cyclist, and I see a clear effect of slower rides in winter. I’m talking about around specific routes and training sessions where the point is to go as fast as possible. With the same power I seem to go more slowly when it’s cold.

Obviously there could be many reasons for this: less aerodynamic clothing, more cautious cornering etc etc, but the effect is very noticeable to me.