Why would an athlete add extra blood to their own body?

Because more red blood cells mean more oxygen reaching the muscles, and more oxygen means an athlete can push harder for longer. That is the entire logic of blood doping: boosting the number of red blood cells in the bloodstream so the body can deliver more oxygen from the lungs to working muscle, raising aerobic capacity and endurance. It can be done by injecting a drug like erythropoietin (EPO) to make the body manufacture more red blood cells itself, by transfusing blood back into the same athlete after storing it, or by transfusing blood taken from someone else entirely.

For a long stretch of its history, this wasn't even against the rules. Blood doping started being used in the late 1960s, mostly by middle and long-distance runners, but it wasn't outlawed until 1986. The first known case happened at the 1980 Moscow Olympics, when Finnish runner Kaarlo Maaninka received two pints of transfused blood before winning medals in the 5,000 and 10,000 meters, perfectly legally. The International Olympic Committee banned the practice in 1985, but for years afterward there was no test that could actually catch it.

That changed in 2004, when a flow cytometry test made it possible to detect blood from more than one person circulating in an athlete's body, by reading markers on the surface of blood cells. Testing an athlete's own stored-and-reinfused blood is harder, and relies on an indirect method involving inhaling a carbon monoxide gas mixture to spot abnormal increases in the blood's oxygen-carrying capacity. None of this is risk-free: adding extra red blood cells thickens the blood, and an overdose of EPO can thicken blood into an artery-clogging sludge, raising the risk of heart attack, stroke, phlebitis and pulmonary embolism.