Suppose there is a little bit of evidence available, such as a red stain on the wall, and one starts to deduce "facts" from that, for example, that someone cut their finger by a knife yesterday morning near the wall (rather similar to Sherlock Holmes), though it can also be the case that someone pierced their finger by a lancet yesterday evening. In other words, rival theories are underdetermined by the available evidence.

Is there any idiom to describe the fact that he is "draining" too much from the evidence "well", or "milking" too much from the evidence "cow"?

  • Not an idiom but. I would use "extrapolating".
    – piccolo
    Commented Apr 12, 2021 at 11:23
  • 13
    ...and I would use overextrapolating. Commented Apr 12, 2021 at 12:32
  • 1
    Extrapolating, in its literal sense, can be applied only to the matters that are expressed quantitatively. It is a respectable form of reasoning, as long as it is applied carefully, and with the awareness of its limitations. Trying to apply that term to the OP's example, however, stretches its meaning considerably.
    – jsw29
    Commented Apr 12, 2021 at 20:10
  • 1
    There's even a whole field dedicated to generating those wild theories from random data! It's called economics. Commented Apr 13, 2021 at 0:53
  • 4
    In statistics, data dredging has this sort of sense. When I was young, data mining meant much the same, but now has a more positive sense. Overfitting is another activity seen as abusing evidence
    – Henry
    Commented Apr 13, 2021 at 15:07

11 Answers 11


Read too much into something

to think of (something, such as a comment or situation) as having a meaning or importance that does not seem likely or reasonable

  • This phrase, as the quoted definition implies, is usually used in the cases in which something is treated as more important than it really is. This doesn't quite fit the OP's example: the cut's being caused by a knife in the morning is not obviously more important than its being caused by a lancet in the evening. The unjustified conclusion in that example concerns the details of what happened, not its importance.
    – jsw29
    Commented Apr 12, 2021 at 16:17
  • @jsw29, the definition mentions "meaning" in addition to importance. Also, you should compare "the cut's being caused by a knife in the morning" with the brute evidence, i.e, there is a red stain on the wall. Commented Apr 12, 2021 at 20:10
  • 4
    This is close, but wrong. We mostly use this phrase for intent. If a cute guy says he just got a new motorcycle and you think he wants to date you, that may be reading too much into it. No one on a detective show ever used this phrase about evidence. Commented Apr 12, 2021 at 21:22
  • 1
    something like "reading too much between the lines"? Commented Apr 13, 2021 at 2:29
  • @OwenReynolds, your motorcycle example is great. The reason that we don't use the term for Sherlock Holmes is that we know he doesn't read "too much into" evidence, rather he observes other evidences as well, and he reads "the" evidences some of which are hidden from us. One who isn't aware of the powers of him may use it. Commented Apr 15, 2021 at 9:55

Jumping to conclusions

Jumping to conclusions ... is ... where one "judges or decides something without having all the facts; to reach unwarranted conclusions".


Jump to conclusions

To make decisions or form opinions before one has all the pertinent facts.

  • I know you found some suspicious things in her office, but don't jump to conclusions—talk to her first.

[Farlex Dictionary of Idioms]

  • 2
    Upvoted, however, there is a caveat: Both have a connotation of "heist" which I didn't mean, or so it seems to me. Commented Apr 12, 2021 at 12:21
  • 5
    @Kaveh, it is not obvious what you mean by 'a connotation of "heist"'; can you elaborate it?
    – jsw29
    Commented Apr 12, 2021 at 16:08
  • 1
    I think I see it. If someone constructs a wall chart complete with yarn arrows and concludes the stain was caused by a knife cut, we wouldn't say they jumped to a conclusion. Commented Apr 12, 2021 at 21:27
  • 8
    @jsw29 Haste not heist. PS But the jump does not imply haste. It says they took a step in reasoning that was too big.
    – philipxy
    Commented Apr 12, 2021 at 22:46
  • 1
    @Kaveh Did you mean haste? Commented Apr 13, 2021 at 4:25

In a case like this I think I'd say:

You put (or added) two and two together, and got five.


Although this expression does suggest that the inference is wrong, rather than only unjustified.


Reading Tea Leaves

This is perhaps a bit more obscure than you're looking for, but it's a form of divination: you make some tea, drink it off, and then look at the pattern left behind by the leaves to infer the answer to some question you have. The expression is often used to describe the process of finding patterns in randomness and possibly using them to explain something unrelated.

  • Citing relevant sources for your answers authenticates them. :)
    – user405662
    Commented Apr 14, 2021 at 0:06
  • 1
    @user405662 yeah, I looked it up originally but forgot to add it by the time I clicked Post. Commented Apr 14, 2021 at 2:00
  • When I've seen this used, it was in situations where you know there is insufficient information to make a definite prediction, but you need to have some idea of what might happen so that you can prepare for it, so you make your best guess. There was no pejorative sense to it as there is in the question.
    – David K
    Commented Apr 15, 2021 at 13:40
  • @DavidK Whenever I've heard the expression used in live conversation, it was a dismissive comment on apophenia. But I'm an engineer, so you can imagine my social circle takes a dismissive view of all forms of divination. Commented Apr 15, 2021 at 18:05
  • I’m trying to remember exactly where I’ve heard this. People in my field don’t literally do divination of any kind. It may have come up in a conference panel where someone asked an unanswerable question about some future conditions, and the answer was couched that way to emphasize that it really was a very uncertain guess.
    – David K
    Commented Apr 15, 2021 at 23:46

Both the aforementioned "reading too much into something" and "jumping to conclusions" are good idiomatic expressions for the meaning sought by the OP.

"Jumping to conclusions" can be expressed less idiomatically and more formally as "faulty generalization". The Wikipedia page cited currently includes some eleven different versions of this, all more or less formal, with the most idiomatic being "leaping to conclusions" and "secundum quid" (where the idiom in question is Latin).

There is also the rather idiomatic expression "making a mountain out of a molehill". This is more typically used to indicate when someone is exaggerating the importance or impact of a small problem, but it can be and is used on occasion to refer to what the OP asks about, drawing too many conclusions from too little evidence.



An opinion or conclusion formed on the basis of incomplete information.

Form an opinion or supposition about (something) on the basis of incomplete information.

Source: Oxford English Dictionary (Lexico link)


In such a scenario I would say "That's a long shot." when I heard their theory for the first time.


The phrase "Take a button and sew a vest on it" expresses this nicely.

(Erle Stanley Gardner was fond of this phrase. I can't recall seeing it elsewhere.)


Not exactly an idiom, but a familiar saying in some circles is "multiplying causes beyond necessity", which in the negative injunction- "Do not multiply causes beyond necessity", is the guiding principle called Occam's Razor. This may not be useful for your purposes, but it is precisely about inferring more than is warranted by the evidence.


An idiomatic and alliterative expression might be:

An inflated inference.


to presuppose

to accept that something is true before it has been proved

Not the answer you're looking for? Browse other questions tagged or ask your own question.