The tropical ichneumonid wasps are poorly sampled, particularly in the Old World, despite their ecological importance and high species richness. Existing inventories also tend to have had a taxonomic focus, with trap placement reflecting the need for maximal sample size rather than rigorous ecological comparisons. Here, we report the results of an intensive sampling effort at Kibale National Park, Uganda, carried out with randomised trap locations. We sampled ichneumonid faunas for a year (2011–2012) with Malaise traps, in eight successional sites ranging from clear-cut exotic plantations to primary rainforest. The traps were situated in 10 random locations at each site and were moved between locations once a week. The total sampling effort encompassed 231 trap months (using traps smaller than the standard size), one of the largest we know of from a single Afrotropical location.We sorted the collected ichneumonids into subfamilies and investigated whether their community composition differed between the sites. Ichneumonid faunas differed between forest and former plantation sites, with both the overall difference and that of four subfamilies significant. Our sample size was unexpectedly small (1212 individuals), but we estimate that the model-based analyses we used could still have given a significant result with a smaller (954 individuals) sample. Overall, randomly placed Malaise traps detected ecological patterns in Afrotropical ichneumonid distributions. Our data also showed that there is a rich and at least partly undescribed ichneumonid fauna still awaiting discovery in the Afrotropical rainforests. Future inventories of this fauna may, however, have to compromise on objective random trap placement in order to get a large sample size.