Ah, that makes it easier to find.
Going off John Jefferson’s The Holy Wars of King Wladislas and Sultan Murad and “The Ottoman-Hungarian Campaigns of 1442″, Mezid Bey had about 16 000 men at Hermannstadt, while Hunyandi’s numbers are unknown. Oruç Bey, an Ottoman historian of the late 15th and very early 16th century gives figures of 10 000 and 15 000 for Hunyandi, depending on which manuscript is used, but these figures are probably inflated.
From the 15th century accounts, it seems clear that Hunyandi’s attack on Mezid Bey revolved around attacking the small number of men he had in the base camp while the vast majority of his army was scattered across the countryside, looting and burning. While he did divide his forces and use the armour switching ruse, the earliest and best accounts having him simply attack the front and two sides of the camp, rather than using part of his force to attack from the rear.
He then dressed some of his soldiers in enemy uniforms and had them play the part of the Ottoman camp guards so that, when the raiders returned, they wouldn’t realise what had happened. This allowed him to hide his army along their approach so that, when they were strung out with their booty and the force in the camp blocked their path, he could sweep down on them and slaughter a large portion of their numbers. Unable to mount an effective resistance because of their massive accumulation of booty and slaves, the raiders proved easy meat for Hunyandi’s soldiers.
For Hunyandi’s second victory, over Şehabeddin (Jefferson puts Mezid at the Iron Gates and Şehabeddin near Silistra), we don’t know the number of Hunyandi’s army1, and the Ottoman forces probably didn’t exceed 44 000 men2. The sequence of events you’ve provided sounds like a fusion of the defeat of Mezid, an unreliable 16th century account of Şehabeddin ‘s defeat and the the Battle of Varna, since Ottoman and early European sources make it clear that Hunyandi allowed the better part of the Ottoman army to cross back over the Danube, primarily the akinji and some of the sipahis, before launching an attack on those left on his side of the river. What followed was a long, bloody fight that lasted until nightfall. Şehabeddin managed to cross back over during the night, along with the janissaries and probably at least some of the sipahis, but the sixteen sanjak beys and the men who stayed with then to cover the crossing were killed or drowned attempting to cross the river after their defeat.
This second victory was necessarily incomplete because Şehabeddin had the good sense to fortify his camp and keep the sipahis and janissaries beyond, while the raiders made their expedition. Hunyandi could only keep a watch on the camp and then, because of the size of the army, only keep up nuisance raids until a large part of it had crossed the Danube.
Coming to the Battle of Varna, we do see a 3:1 force disparity, with the Crusaders most likely numbering only 16 000 compared to somewhere between 50 and 60 000 Ottoman soldiers. The larger Ottoman numbers allowed them to outflank the crusaders on the right (a risky move since terrain would prevent the Rumelian troops from retreating), but terrain selection prevented the crusader left from being outflanked.
The Ottoman left was able to crack, if not completely shatter, the crusader right, but the wagons placed at the rear prevented a rapid break out and attack on the rear of other parts of the line, allowing for Hunyadi to bring up part of the reserve, which was then beaten back. However, the superior armour of the crusaders on the left stalled the attack of the Ottoman right and, when their commander was killed, the Ottoman right wing broke and fled. This allowed the crusaders on the left to aid those on the right, forcing the Ottoman left into flight.
The Ottoman center, however, contained the best troops, and these had been behind quite considerable field fortifications up until this point. The Sultan now threw them into the battle. As skilled, well equipped and fresh as they were, the Ottoman slave troops were unable to defeat the crusaders, who may have outnumbered them3. They were driven back to their fortified camp, where King Wladislas made an ultimately suicidal attack with no more than 500 men in an attempt to strike a killing blow. Hunyadi retreated during the night, while some crusaders, thinking they had lost the battle, also retreated in a different direction. The remaining crusaders retreated to the wagons, while the Ottoman forces slowly coalesced around the Sultan’s camp as word spread of Wladislas’ death.
The key to the relative success of the crusaders during Varna, in spite of being outnumbered, seems to me to be a combination of superior armour compared to the Ottoman levy cavalry and choosing a site where it was very difficult for the Ottoman troops to outflank them. Using the wagons as a barrier to their rear and goading the Ottomans into attacking so that they didn’t just starve the crusaders out were also important contributing factors to their early success in the battle.
1 A letter from King Wladislas, preserved in an Italian chronicle, gives Hunyandi 25 000 men. This is likely an exaggeration, since only 14-20 000 men (and most likely the lower figure) were mustered by Hungary for the first stage of the Varna Crusade.
2 Based on the average number of sipahis of the Rumelian provinces being 1156, the average number from the Anatolian provinces being 1820 and there been ten Rumelian sanjak beys from and six Anatolians. This provides a total of ~25 000 sipahis, 15 000 akinji and 2-4 000 janissaries. The sipahis, it must be noted, were of very variable quality and some were armed with no more than a wooden club, while others were equipped better than any European knight.
3 Casualty figures are not available, and it’s unknown how many crusaders had pursued the fleeing Ottoman cavalry. I think it’s plausible, however, than as many as 10 000 remained, compared to 6-8 000 Ottoman slave troops.
