Background The main objectives of this study were to evaluate the accuracy of different body weight formulas for estimating body weight of Icelandic and Warmblood horses, as well as to assess the associations between the variables cresty neck score, body condition score, and plasma concentrations of leptin, insulin and cortisol. A total of 81 adult (≥ 4 years of age) horses (43 Icelandic and 38 Warmblood horses) was included in this study. The following morphological measurements were collected by two examiners simultaneously; body weight; height at withers; neck length; 0.5 neck length; neck circumference; umbilical circumference, two different heart girths, as well as two different body length measurements. The horse’s body weights were measured on a weight scale, and cresty neck scores were rated along with body condition scores using both the 0 to 5 and the 1 to 9 systems. Results In general, the concordance correlation coefficient was high for most formulas, but the mean bias and slope bias deviations varied between formulas. Some simple formulas using only heart girth, or heart girth and length can be used to estimate body weight of Icelandic and Warmblood horses as good as more complex formulas using four morphometric measurements. Plasma concentrations of leptin and insulin were higher (P < 0.001) for the Icelandic than the Warmblood horses, probably reflecting higher body fat content as suggested by the differences in body condition score. Conclusions Body weight formulas only give an estimate of body weight and not a completely correct determination. Some simple and more complex formulas can be used for Icelandic horses even though they are not developed for this breed. Complex formulas using four morphometric measurements were accurate, but simple formulas using only heart girth, or heart girth and length can be used to estimate body weight and thereby be applied to weight tapes and used to estimate the body weight of both Icelandic and Warmblood horses.