Ok the answer is 2.4%. This means they win (2.4/1.9)/1.9= 26.3% more races than they should. This is what is important!

A stat that can stand alone as pointing out that all other things being equal, more horses that have this characteristic win, than should win.

If there were 1000 races run with 12000 runners the numbers would be this:

.019*12000= 228 runners
.024*1000=24 winners
SR= 24/228 = 10.5%
To break even you would need an average dividend of $9.52

At this stage it is worthwhile pointing out that if the 12 runners in each race had an equal chance the strike rate would be 8.5%

It should also be noted that there is no single factor which can be blindly bet on every race every day to show a profit. (at least none that I am aware of)

In fact I think you might find that if you only wish to consider one thing the strongest single factor (ie more winners than expected by random luck), is favourites. You'll have a good strike rate but lose about 12c in the $ year in year out but its reassuring to know that everyone else agrees with you - even if you are wrong 70% of the time.