Some highly irradiated close-in exoplanets orbit stars showing anomalously low stellar chromospheric emission. We attribute this to absorption by circumstellar gas replenished by mass loss from ablating planets. Here we report statistics validating this hypothesis. Among ~3000 nearby, bright, main sequence stars ~40 show depressed 51chromospheric emission indicative of undiscovered mass-losing planets. The Dispersed Matter Planet Project uses high precision, high cadence radial velocity measurements to detect these planets. We summarise results for two planetary systems (DMPP-1 and DMPP-3) and fully present observations revealing a Mp sin i = 0.469 MJ planet in a 5.207 d orbit around the γ-Doradus pulsator HD 11231 (DMPP-2). We have detected short period planets wherever we have made more than 60 RV measurements, demonstrating that we have originated a very efficient method for detecting nearby compact planetary systems. Our shrouded, ablating planetary systems may be a shortlived phase related to the Neptunian desert: i.e. the dearth of intermediate-mass planets at short orbital periods. The circumstellar gas facilitates compositional analysis; allowing empirical exogeology in the cases of sublimating rocky planets. Dispersed Matter Planet Project discoveries will be important for establishing the empirical mass-radius-composition relationship(s) for low mass planets.Massive planets in short period orbits were the first exoplanets to be discovered, but their low-mass analogues remained elusive for almost two decades. Recently, however, Kepler results have established there is a significant population of low-mass, ultra-short period (USP) planets: 0.51 ± 0.07% of G dwarfs host low-mass USP planets with Porb < 1 d 1 . Most of these planets are smaller than 2 R⊕. Kepler has a narrow field of view, so the planets it discovers are generally distant and difficult to study in detail. Radial velocity (RV) mass determinations of the overwhelming majority of Kepler's small planet discoveries are consequently more-or-less precluded. Nearby analogues of these Kepler USPs must exist, but an enormous effort would be required to find them through targeted RV observations of bright nearby stars with only a 0.5% chance of success for each target. DMPP selects target stars likely to host planets with short orbital periods using archival stellar spectral information. This paper (i) explains the motivation and underlying hypothesis for the DMPP selection criteria, which were developed using inferences from observations of transiting hot Jupiters; (ii) uses the statistics of our first three DMPP planetary system discoveries to test the underlying hypothesis (iii) presents the discovery of DMPP-2 and our RV survey methodology. Companion papers present the discoveries of DMPP-1 and DMPP-3 (this issue of NA); the characteristics of these two systems are briefly summarised to place the hypothesis testing in context.The discovery of an extensive H I exosphere surrounding the hot Jupiter HD 209458 b 2 , and subsequent evidence suggests th...