Typical values are generally an average from a representative number of batches. Specifications are an agreement between the supplier and customer with consequences for missing the limit.
Specifications are a range that encompass typical values and are therefore broader in scope. The gap between typical and specification values takes into account test reproducibility between labs, batch-to-batch variations, and an additional margin to avoid liability with the customer.
Test reproducibility alone is often a large part of the gap. The pour point test under ASTM D 97 is not very precise. The reproducibility between labs is 6°C or 11°F. So if one lab gets -54°C and another gets 48°C, the results are considered an acceptable match. Add to that the batch-to-batch variations and some safety margin and it is easy to see why a product with a typical value of -54°C could have a specification limit of -45°C max.
In addition, not every company choses an average to report their typical values. Some can be a bit liberal in their favor, especially in a highly competitive market where such numbers are compared, and some a bit more conservative. Combined with the factors above, this is why using pour point results to guess at an oil's base oil composition is far from accurate. The same is true for many other tests.
From ASTM D97-17b (Pour Point)
"Reproducibility—The difference between two single and independent test results, obtained by different operators working in different laboratories on identical test material, would in the long run, in the normal and correct operation of this test method, exceed 6°C only in one case in twenty. Differences greater than this should be considered suspect."
