Your logic is deeply flawed. For a couple different reasons.
1. Intimate partner violence (IPV) as defined by the CDC is actually a pretty broad spectrum of experiences that is not limited to people the victim is currently in a romantic relationship with (Link to their actual definition below, which you’ll note is not from wikipedia — on a related note wikipedia isn’t the best source to go to for backing up an argument) A person could be victimized by a close friend and still be included in the IPV statistics. So really the only thing you’re data says is that women, and in particular lesbians are more likely to be victimized then men, which is unrelated to your argument that men are not more violent then women.
2. More importantly. Even assuming the IPV statistics referred exclusively to violence between people in romantic relationships, you’re argument would still be objectively incorrect. This is because the percentage of the population that is gay and lesbian is vastly smaller than the percentage of the population that is heterosexual. The General Social Survey (Link directly to the dataset below) shows that 1.7% of the population identifies as gay or lesbian compared to 96.1% identifying as straight. These numbers are consistent with other surveys on the topic which you are welcome to research.
I’m going to further show why you’re argument is flawed using some basic math. Let’s say we have 1000 women(for the sake of making this math simple). Based on the above statistics, 170 of these women would be lesbians and 961 would be straight women. Using the statistics you cited (and assuming they referred exclusively to victimization by a romantic partner) within the lesbian population 75 (43.8% of 170, rounded up) women would be statistically likely to be victimized by their romantic partners, thus 75 women in this sample would be abusing other women in lesbian relationships. In the straight population of women 337(35% of 961, rounded up) women would be statistically likely to be victimized by their romantic partners giving you 337 violent men in this sample.
337 is larger than 75, because, duh.
You would have been better off pulling data about acts of violence in general committed by men vs women (except you’d still see that men commit more, as is well referenced in the original piece you called “flawed”). But basing a counter on sexual orientation is actually absurd.
2A: In order to see the data about sexual orientation specifically type “sexornt” into the “Variable Selection” line and click “View”