Morphy v The Petrov (conclusions)

PAUL Morphy faced the Petrov six times, scoring +3=2-1 (66.7%).
On the face of it, that was a poor percentage for him, but the strength of the opposition, and the nature of the games, including blindfolds and simuls, should be taken into account.
In addition, Morphy failed to convert probably-winning endings in both the draws, and the loss was a wild game that could have gone either way.
The first time against the Petrov, he played the Cozio Attack (3.Nxe5 d6 4.Nf3 Nxe4 5.Qe2), but seems to have concluded that White's edge was too small (this is in line with modern thinking about this system).
In his next two outings, he played the Boden–Kieseritzky Gambit (3.Bc4 Nxe4 4.Nc3).
Despite winning both games, he switched to the Classical Attack (3.Nxe5 d6 4.Nf3 Nxe4 5.d4), which Johann Löwenthal had played against him three times in their 1858 match.
Those three games ended in a win apiece and a draw, and it is tempting to think this made an impression on Morphy.
The trouble with this theory is that the two Boden–Kieseritzky Gambits came either side of the match.
Morphy scored a win and a draw with his two Classical Attacks, but for his final outing against the Petrov, he switched back to the Boden–Kieseritzky Gambit.
This was the only game of the six that he lost, so it is anyone's guess what he would have played had the question arisen again.
What can be said from all six games is that, as against the French, Morphy strove for quick development in an open position as a prelude to an attack on the opposition king.