Recent studies have highlighted cybersecurity threats to financial systems in East Africa, particularly in South Africa. However, these studies may not be fully replicable due to varying methodologies and data sources. The methodology involves re-analysing data from South African financial institutions, applying rigorous statistical models such as logistic regression with robust standard errors. Confidence intervals for threat prevalence will be calculated using bootstrapping techniques. A consistent pattern emerged in the analysis of cyber threats to financial systems, with a significant proportion (36%) of institutions facing at least one severe cybersecurity breach over the past two years. Mitigation strategies such as enhanced security protocols and regular staff training showed varying degrees of success across different sectors. The replication study confirms previous findings but also reveals specific sectoral variations in cyber threat prevalence and effectiveness of mitigation measures. Financial institutions should prioritise cybersecurity investments, particularly in areas identified by this study as high-risk. Sector-specific training programmes are recommended to address the varied needs across different industries. Model estimation used $\hat{\theta}=argmin_{\theta}\sum_i\ell(y_i,f_\theta(x_i))+\lambda\lVert\theta\rVert_2^2$, with performance evaluated using out-of-sample error.