In Defense of the Mystique Despite the insightful analysis of my friend and mentor Dr. Alan Nathan, and as a former pitcher, current catcher and erstwhile mathematician, I hereby rise to sprinkle more fairy dust on the mystique of the anecdotal knuckleball. One of my current battery-mates is a former major-league pitcher and can throw a decent knuckleball. And heeding Uecker’s advice, I occasionally “ … wait ‘til it stops rolling and pick it up.” Of course this may say more about me than about the knuckleball. Another battery-mate - who pitched in Baltimore’s AAA system - shares the following anecdote which he witnessed (yeah he’s way old – like me): (And here I paraphrase –) ‘Spring Training 1959; Gino Cimoli (Cards) faces Hoyt Wilhelm (O’s). Gino swings at a pitch and misses – it hits him square in the gut – Strike 2. Next pitch – same thing – a gut-buster – Strike 3. Gino returns to the dugout with his doubly-bruised gut and flagging ego.’ Whatever you may think of Gino, his major-league batting eye did fairly well in the subsequent season in 1959: .279 in 569 PA. Orioles catchers went on to set an MLB record with 49 passed balls in 1959. And it was in the next season that Paul Richards introduced the 45” Large Mitt to protect Clint Courtney and Gus Triandos from Hoyt Wilhelm. Something baffling must be going on here. Perhaps the knuckleball is a moving optical illusion set up by the unexpected frozen image of the approaching seam pattern. And as another Commenter has suggested, perhaps the answer – and the solution - lies more within the world of human neurons, ganglions and photoreceptors than mystique-al pitches … Please hurry - I sure do need help catching the knuckleball … On a more – or less – serious note, please consider the second figure in Alan’s paper. Assuming my less-than-complete understanding of PITCHf/x references, and based on the movement pattern shown in the upper plot (presumably an RHP curve ball), it appears to me that the figure should be labeled “trajectories as viewed from below” (as if up through a glass floor) rather than “above” - as awkward as this might at first seem. A small correction involving a rotation of the ordinate axis or reflection of the plotted data - might be in order. I think Alan is already considering this.