> iBuyers are always low-balling Low-balling according to their own estimate (Zestimate) with imperfect knowledge. If this Zestimate overvalues a house, but th

> iBuyers are always low-balling Low-balling according to their own estimate (Zestimate) with imperfect knowledge. If this Zestimate overvalues a house, but the iBuyers assume “we are low-balling anyways,” they may find themselves in an iBidding war ending in the winner’s curse: whoever pays the most for the overvalued house wins. Not a problem though right? They overvalue some, undervalue others, it evens out. But, as the article points out, the nature of this algorithmic buying is that the iBuyers will only - or at least predominantly - end up with the houses they overpaid for or got only a slightly good deal on because many sellers will be willing to take longer for a better deal. The end result is very thin margins or, as seen with Zillow, substantial losses as they hold a lot of lemons.