N size for not being considerated leverage - r

Question

Suposse I have a dataset with two variables x and y, with the purpose to run a linear regression y ~ x.
We have all x values equal and y varying between 1 and 10.
For example (in R code):
x <- rep(100, 50)
y <- runif(50, 1, 10)
If I add a new value, being x value 75, this new value will be considerated a leverage:
x <- c(x, 75)
y <- c(y, runif(1, 1, 10))
fit <- lm(y ~ x)
im <- influence.measures(fit)
tail(im$is.inf)
How many 75's I need to add to the dataset for not being considerated a leverage?
Is there any R package that returns that critical N size?
Edit after #RuiBarradas comments
hatvalues with 51 observations (50 100's and 1 75) are:
> im$infmat[, 6]
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02
17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02
33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02
49 50 51
0.02 0.02 1.00
What i want to know is how many 75's i have to add so that 75 are not considered leverages, because I'm deleting observations with high leverage from my analysis.
This have to be done programatically for over 1000 cases.

Answer 1

Well I don't think anyone is trying to question your approach just point out that as you drop high leverage rows more will appear. potentially with even higher leverage.
But best to simply arm you to do what you want. Here's a function that will take a dataset, a regression formula, and how_deep you want to go.
We'll apply it to mtcars and say that we want to make sure we have 27 rows of 32 left which means how_deep = 27. You get back a vector of cars. Initially the Lincoln has the highest leverage when you eliminate it, it's now a Chrysler with what is actually higher leverage. Hopefully your data will be much different
recursive_leverage <- function(data, formula, how_deep) {
while (nrow(data) >= how_deep) {
data$hatval <- hatvalues(lm(formula = formula, data = data))
print(paste(rownames(data[which.max(data$hatval),]), max(data$hatval)))
data <- data[-which(data$hatval == max(data$hatval)),]
}
}
recursive_leverage(mtcars, mpg ~ wt + disp, 27)
#> [1] "Lincoln Continental 0.198891254526384"
#> [1] "Chrysler Imperial 0.240276382749128"
#> [1] "Cadillac Fleetwood 0.274528186610731"
#> [1] "Lotus Europa 0.224979770471144"
#> [1] "Honda Civic 0.223347867723116"
#> [1] "Ford Pantera L 0.21141405715464"