Anchor tests play a key role in test score equating. We attempt to find, through theoretical derivations, an anchor test with optimal item characteristics. The correlation between the scores on a total test and on an anchor test is maximized with respect to the item parameters for data satisfying several item response theory models. Results suggest that under these models, the minitest, the traditionally used anchor test, is not optimal with respect to anchor-test-to-total-test correlation; instead, an anchor test with items of medium difficulty, the miditest, seems to be the optimum anchor test. This finding agrees with the empirical findings of Sinharay and colleagues that the miditest mostly has higher anchor-test-to-total-test correlation compared to the minitest and mostly performs as well as the minitest in equating.