I would be very happy with 70% recoveries for most surrogates by 8270.
2,4,6-Tribromophenol 40-87
2-Fluorobiphenyl 18-97
2-Fluorophenol 35-79
Nitrobenzene-d5 24-150
Phenol-d6 22-95
Terphenyl-d14 35-128
These are more typical limits which have been derived over time using the statistical formula listed in the method using recoveries in ongoing samples. (several hundred used each time they are evaluated)
Those are for water samples, soils can be even lower depending on technique.
If you are running just blanks, recoveries can be in that 70-130 range, but even then the 70% can be difficult to hit for some surrogates due to loss in concentration or extraction efficiency. Real world samples with dirty matrix normally pull those recoveries down so that is why they suggest using the statistical method to set ranges to better represent what you would find in real world samples.
8260 usually has better recoveries since normally the internal standard and surrogate are in the same mixture when injected, especially if using automated addition in the autosampler. For 8260 waters our recovery limits are in the 80-120 range but for soils the BFB drops into the mid 60%s on the lower end. Any soil with clay in it can adsorb it quite well.