Cobalt oxyhydroxide (CoOOH) has attracted great attention in electrochemical water splitting. However, the mechanism behind its catalytic performance and how to improve its activity are still under debate. In the work, we propose that strain engineering is an effective and simple way to achieve the purpose. Based on density functional theory (DFT), we investigate the effects of strain engineering on the electronic structure and catalytic performance of two-dimensional (2D) CoOOH and the underlying mechanism of the oxygen evolution reaction (OER). We find that strain engineering is effective to tailor the electronic configuration of Co3+ ions in CoOOH, which can be transferred from low spin (LS: t62ge0g) to high spin (HS: t42ge2g) at a tension of 9%. Importantly, we show that LS CoOOH is a poor OER catalyst, because it is ineffective for O2 release with a large energy (1.35 eV). However, HS CoOOH is much more active in the OER because of smaller O2 release energy (0.03 eV) and more effective O–O bond coupling (1.21 eV) in the intramolecular oxygen coupling mechanism. The overpotential for LS CoOOH is 0.66 V according to the hydroxide ion attack mechanism, while the direct intramolecular coupling is hard to occur. HS CoOOH shows low overpotentials, 0.32 and 0.5 V, for the intramolecular coupling and hydroxide ion attack, respectively, which are comparable to those of the best OER catalysts (0.25 to 0.4 V). Our work demonstrates that the spin state transition of Co3+ ions tuned by strain engineering is an effective way to improve the OER activity of 2D CoOOH.