We collected some data several years back.
- serials started at 1000.
- about 1/2 of the 1920 were made in the first year (around serial 53xx).
- 260remguy (Jeff) who had collected 1920/20/26's said the earliest 300 chambering he had heard of was 39xx.
- after that point (39xx) my data shows about one .300 to every two in .250 till the design change.
- as for the Model 20 (serial 105xx) it looks more one for one.
The Model1920 was designed for the lighter recoiling .250.
The Model 20 had a heavier stock and handled the .300 better.
Highest serial # 13006
I think that figures around 70% .250's to 30% .300's
I have been collecting, or at least accumulating, 1920s and 20/26s for over 40 year.
I have owned 4 1920s in 300 Savage, the lowest SN was 39xx and the highest was 74xx.
I have owned over 40 1920s in 250-3000, the lowest SN was 1015 and the highest was 101xx.
I have owned 26 20/26s in 300 Savage, the lowest SN was 106xx and the highest was 129xx.
I have owned 24 20/26s in 250-3000, the lowest SN was 105xx and the highest was 125xx.
According to information that I received from Mr. Clark, 105xx was made on the first day of production for the 20/26 configuration in July 1926.
Based on the numbers that we have and feel are accurate, there were approximately 9500 rifles made in the first/original 1920 configuration and 2500 rifles made in the second/improved 20/26 configuration.
When I lived in New England and was active in the market, the 1920s were significantly more common in 250-3000, while the 20/26s were significantly more common in 300 Savage.
EDIT: You never know where you'll find a Savage 1920 or 20/26. I found SN 1015 at a crappy little gun show in Leadville, CO. It is very worn, almost no bluing left and the checkering has been worn smooth. It is so worn that if hadn't been for the very low SN, I wouldn't have bought it, but how could I resist?